Sociological Studies of Children and Youth : Special International Volume (Sociological Studies of Children and Youth)

EDITORIAL BOARD

Patricia A. Adler University of Colorado

Gary Alan Fine Northwestern University

Peter Adler University of Denver

Sandra L. Hofferth University of Maryland

Spencer E. Cahill University of South Florida

Jens Qvortrup Norwegian Centre for Child Research

William A. Corsaro Indiana University

Barbara Schneider University of Chicago

Donna Eder Indiana University

Alford A. Young, Jr. University of Michigan

ix

CONTENTS

EDITORIAL BOARD

ix

INTRODUCTION: INTERNATIONAL PERSPECTIVES ON CHILDREN AND YOUTH Loretta E. Bass

1

WORKING CHILDREN IN ZIMBABWE Michael F. C. Bourdillon

7

TOYS AND GAMES: CHILDHOOD IN THE PARQUE DAS ˜ NAC ¸ OES FAVELA IN BRAZIL Ethel Volfzon Kosminsky and Laura Daniel

23

TRANSITION TO ADULTHOOD IN JAPAN AND KOREA: AN OVERVIEW Hyunjoon Park and Gary D. Sandefur

43

ETHNIC IDENTITY AND SEGMENTED ASSIMILATION AMONG SECOND-GENERATION CHINESE YOUTH Harry H. Hiller and Verna Chow

75

THE RADICALIZATION OF THE SELF – “BEYOND” GENERATIONAL ORDER: GERMAN CHILDREN’S LITERATURE AS A CASE STUDY Doris Bühler-Niederberger

vii

101

viii

CHILDREN’S VIEWS ON CHILDREN’S RIGHTS. A COMPARATIVE STUDY OF SPAIN AND ITALY Angelo Saporiti, Ferran Casas, Daniela Grignoli, Antonio Mancini, Fabio Ferrucci, Marina Rago, Carles Alsinet, Cristina Figuer, Mònica González, Mireia Gusó, Carles Rostan and Marta Sadurní

125

INVOLVING CHILDREN IN SOCIAL POLICY: A CASE STUDY FROM NORTHERN IRELAND Madeleine Leonard

153

THE GENERATIONING OF POWER: A COMPARISON OF CHILD-PARENT AND SIBLING RELATIONS IN SCOTLAND Samantha Punch

169

A TALE OF TWO CITIES: HEALTH-COMPROMISING BEHAVIORS BETWEEN HUNGARIAN AND AMERICAN YOUTH Kevin M. Fitzpatrick, Bettina F. Piko, Darlene R. Wright

189

MAPPING CHANGES IN THE SOCIAL LIVES OF CHILDREN: AN ANALYSIS OF ACTIVITIES AND BEST PRACTICE INITIATED THROUGH SCHOOL-BASED MENTORING IN PHILADELPHIA Suellen Butler

213

THE IMPACT OF WAR, ADULT HIV/AIDS, AND MILITARIZATION ON YOUNG CHILDREN’S MORTALITY Steve Carlton-Ford

231

CHANGES IN NONMARITAL COHABITATION AND THE FAMILY STRUCTURE EXPERIENCES OF CHILDREN ACROSS 17 COUNTRIES Jeffrey M. Timberlake and Patrick Heuveline

257

ABOUT THE AUTHORS

279

INTRODUCTION: INTERNATIONAL PERSPECTIVES ON CHILDREN AND YOUTH Loretta E. Bass This International Volume of Sociological Studies of Children and Youth shows the breadth of empirical research that focuses on children and youth around the world. Across these articles arranged by region, it becomes clear that we assume different ideas about what childhood is even though these are bound by both cultural and structural factors. We often take “children” or “youth” as a definitive given, and then seek to solve their problems or create policies that serve them. Rarely do we have the luxury of actually thinking about the meaning of these two words. This annual volume creates a space for this particular dialogue to take place. Across these research papers, cultural expectations influence how societies view children and how children view themselves. Immigrant children and youth provide particularly interesting insight as they navigate more than one cultural context and varying expectations for children as they negotiate who they are as individuals and children. Structural factors also become salient, as children come from unequal backgrounds and different levels of economic development, and face varying political concerns. Bourdillon provides the one empirical contribution centered on Africa. In his paper, he provides insight into the lives of working children in Zimbabwe. Using data from interviews with children and their families, he discusses both the value and costs of children’s work. The vignettes he presents illustrate the need to understand children within a particular context. Unless this need is met, sensible Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 1–6 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10001-9

1

2

LORETTA E. BASS

strategies to address this pressing social issue cannot be devised. While his findings are based on fieldwork from one country in Southern Africa, his conclusions hold true for most countries in sub-Saharan Africa. African countries rank among the poorest in the world and sub-Saharan African countries have the highest rate of working children, 29%, compared with 19% of children working in Asia and the Pacific, 16% of children working in Latin America and the Caribbean, and 15% of children working in the Middle East and North Africa (ILO, IPEC, SIMPOC, 2002, see also Bass, 2004). From South America, Kosminsky and Daniel examine the “socially excluded” children who live in the Parque das Nac¸o˜ es favela in Sao Paulo, Brazil. Using both observations and interviews with children, they report that children in this economically disadvantaged situation are aware that there are economic and social hierarchies and that they are at the bottom of them. The authors show that regardless of their depressed social situation, these children remain resilient in their childhood creativity, because they play, make toys and socialize in a manner that is unique to childhood. The authors contend that even though these children reside in a neighborhood and hold a social position that may deny them minimum human and civil rights, they remain distinctive in their actions that define them as children. Considering Asian children, Park and Sandefur examine the transition to adulthood among Japanese and Korean youth, and do so by comparing their transitions to that of American youth. The authors take a social demographic perspective and use published data sources to profile the transition to adulthood along four factors: the completion of education, entering the labor force, family formation, and leaving home. They find that strong linkages across education, marriage, and parenthood still exist in among Japanese and Korean youth, compared with their American counterparts. Even though Asian governments invest less in public education compared with the U.S., Asian families choose to invest considerably more in their children. The authors also find that Korean and Japanese youth leave home, marry and have children at later ages, compared with American youth. The authors suggest both institutional and cultural forces influence these divergent outcomes across Asian youth and American youth. Hiller and Chow examine ethnic identity and segmented assimilation among second-generation Chinese youth in Canada. Specifically, they examine whether segmentation exists and how members of an ethnic group understand themselves vis-`a-vis others in society. While prior research has focused on socio-economic factors as a measure of assimilation, the authors focus on socio-cultural factors as expressions of identity in order to understand how identity is related to roles and social location in society. They ask how second generation Chinese youth understand and negotiate their identities, and how this process is related to the process of segmentation. They find that youth divide personal identity into “private

Introduction

3

and public spheres.” If youth feel comfortable or secure in their ethnic identity in the private sphere, meaning their family life, then they are more likely to explore the public sphere while resisting a segmented outcome. Beyond the ethnic group studied by the authors, their findings have implications for children and youth migrants around our world as people move in large numbers across borders. From Europe, B¨uhler-Niederberger turns the concept of the “social self” from socialization theory into a constructivist concept by noting the absence of generational order in our current thinking on children. Using excerpts and content analysis techniques from six children’s books, she evaluates how childhood is conceived and negotiated. Positing that children’s literature is both guide and mirror, she finds that self-reflection, self-control, and self-projection all inform the development of the “self” for children and that these influence levels of authority across generations. She argues that the socialization of the self is to a large extent, self socialization, as children reflect, control, and project themselves, and that this socialization is embedded within a generational arrangement. Saporiti et al. examine children’s perspectives on the various aspects of children’s rights. The authors first provide a review of other studies focused on children’s views of their rights. They test whether cultural and structural factors – namely age, gender, culture, as well as social context and living condition – influence children’s awareness and thinking about their rights. Using data collected in Catalonia, Spain and Molise, Italy, they report minor dissimilarity between children from Molise and Catalonia, between boys and girls, and across children of different ages. Aside from age, however, they maintain that these differences are not as expected. They further argue that our children appear to live in a “flat world” where little makes a difference regarding their views on children’s rights. From these findings, the authors put forth some tentative hypotheses to inform future research on this important issue. Leonard presents a case study from Northern Ireland that examines how children are involved in social policy. She first introduces and then applies Hart’s (1992) influential “ladder of participation” model to children taking the Eleven Plus test in Northern Ireland. The Eleven Plus test was taken by a majority of 10- and 11-year-old children in order to assess what type of secondary-level education would be appropriate for them. Leonard included the voices and perspectives of children on this test so that policymakers can hear and consider their input. Leonard found that Hart’s ladder of participation provided a useful model for including children’s views in the research project, but she also found some difficulties that her paper examines in more detail. Punch compares children’s sibling relations in her study of children in Scotland. Using empirical data gathered of 90 children from ages 5 to 17 and 30 focus groups of siblings, she examines the generational power structure present in families from

4

LORETTA E. BASS

the perspective of children. She finds that child-parent and sibling relationships both exhibit power struggles and negotiations, but children are more likely to cooperate with parental power because they hold authority over children through cultural norms and through control of household resources. She finds that siblings do not exhibit the same degree of disciplinary power as that of parents. Punch uses the concepts of Goffman to consider why the interactions with parents are more self-disciplined and restrained. She finds that children do not feel a similar requirement in their “presentation of self” while engaging in sibling interactions. Punch maintains that this generational difference in power can free siblings to interact more openly with one another. Fitzpatrick et al. examine health-compromising behaviors using 1,242 Hungarian and 1,538 American youth, each residing in an urban area. Specifically, they sought to understand the differences in cigarette smoking, alcohol consumption, illicit drug use, weapon possession, and fighting across their two samples. They find that African-American youth carry weapons, fight, and smoke marijuana more than Hungarian youth. They also show that Hungarian youth smoke cigarettes and drink more alcohol than African-American youth. While they engage in different types of risky behaviors, both samples find that deviant group membership, depression, family violence, and personal attitudes toward drug and alcohol use all co-vary with health-compromising behavior. Across both samples, the authors find that academic achievement and parental supervision protect youth from health-compromising behavior. The authors assert that their findings provide evidence that “risk and protection may operate independent of cultural specificity” in predicting health risks for youth. Focusing solely on North America, Butler explores the activities and practices of school-based mentoring in Philadelphia. The author maps changes in the social lives of children using data gathered through observations and interviews gathered over a six-year period, 1995–2000, while the author observed the AMERICORP mentoring programs in Philadelphia’s public schools. Analyzing tutoring programs, socialized recess and after-school club, the author examines the types of changes these mentoring programs create for children and their schools. In conclusion, the author asserts the potential of these mentoring programs to reconfigure the social life of children. Using a comparative perspective, Carlton-Ford examines the impact of war, adult HIV/AIDS, and militarization on children’s mortality. His study offers insight to the larger question of the impact of war on civilians around our world. Using a sample of 88 countries, the author finds that armed conflicts increase young children’s mortality rates even after controlling for potential sources of spuriousness, such as religious composition of a country, as well as major mediating and moderating factors, such as access to safe water and the interaction of militarization and adult

Introduction

5

HIV infection rates. His results point to a direct relationship between militarization and the health of children and to civilian populations more generally. Timberlake and Heuveline again draw on comparative data to examine cohabitation and the experiences of children across 17 industrialized countries. Using data from the Family and Fertility Surveys, they estimate the effects of adult cohabitation, in terms of incidence and duration, on the family structure experiences of children. They find that children are increasingly exposed to cohabitation even though much of the increase in non-marital cohabitation has occurred when children are not residing in that household. This study finds that changes in the prevalence of cohabitation has affected children some but probably not as much as commonly held. The authors maintain that the prevalence of cohabitation is likely to become a strand of what demographers are calling the “second demographic transition.” This transition will change family structures as children know them among Western industrialized countries. Each of these articles provides insights. Much like the larger interdisciplinary field of childhood studies, these articles come from a range of theoretical and methodological orientations, yet all offer a child-centered approach. International childhood researchers push our paradigms into unlikely contexts. This exposes the limitations of our thinking on children and youth studies, yet forces our theoretical models to be more versatile and robust. The diversity within studies of children and youth provides a rising tide that can lift all boats for children’s issues. While these papers come from different doorsteps of the world, there are threads of continuity that connect them as meaningful for children, youth and the emerging field of childhood studies. In particular, the research of B¨uhler-Niederberger and Punch reasserts the importance of generational relationships and the implications of these relationships. The research of Bourdillon, Kosminsky and Daniel, and CarltonFord underscores the salience of political and ecomonic factors affecting children and their families. The papers of Bourdillon, Kosminsky and Daniel, Saporiti et al., Leonard, and Butler underscore the significance of children’s voices in debates and programs affecting children. The papers of Park and Sandefur, Hiller and Chow, Timberlake and Heuveline, and Fitzpatrick et al. highlight the gains that can be made by comparing structural and cultural factors across societies. All of these papers make unique contributions yet at the same time they build bridges across continents and issue areas. These papers also show how some forces affecting children remain local while others are global. In closing, I thank the regular Series Editors, David A. Kinney and Katherine Brown Rosier, for their invitation to edit this Special International Volume of Sociological Studies of Children and Youth. I with to thank the review board and anonymous outside reviewers who provided detailed comments on the papers submitted for this volume. I thank all who submitted papers and I thank those

6

LORETTA E. BASS

authors whose papers are included in this volume. I could not have asked for more cooperative and patient colleagues. Thank you all.

REFERENCES Bass, L. (2004). Child labor in Sub-Saharan Africa. Lynne Rienner Publishers. Hart, R. (1992). Children’s participation from tokenism to citizenship. Florence: UNICEF. ILO, IPEC, SIMPOC (2002). Every child counts: New global estimates on child labour. Geneva: International Labour Organization (ILO), International Program on the Elimination of Child Labour (IPEC), and Statistical Information and Monitoring Program on Child Labour (SIMPOC).

WORKING CHILDREN IN ZIMBABWE Michael F. C. Bourdillon INTRODUCTION This chapter is about the value and costs of the work of children – to themselves, their families, and their communities. I present a range of situations in which children in Zimbabwe have to work, placing the employment of children in the broader context of children’s work. I point to benefits and harm that can accrue to children from their work, paying attention to how the children themselves perceive it. The vignettes I present illustrate the complexity surrounding the work of children, and the need to understand the contexts in which they work before we try to intervene. The international campaign against child labor has become increasingly effective in recent years. Within three years, 131 countries had ratified the convention of the International Labour Organisation on the worst forms of child labor (Convention 182, 1999; ILO, 2002b, p. 19). A variety of non-governmental organizations have supported the campaign, and over 80 countries have associated themselves with the ILO’s International Program for the Elimination of Child Labour. Evidence suggests an overall drop between 1995 and 2000 in the proportion of children working (ILO, 2002a, pp. 19, 20). At the same time, people are becoming increasingly aware of the complexities surrounding working children. In appropriate situations, work and even employment, can be beneficial to children. My introduction to child labor in Zimbabwe was in schools in which children worked long hours on tea estates in order to pay for their schooling, and sometimes also to contribute to the livelihood of their families. On the one hand, the situation of the children was very hard and Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 7–21 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10002-0

7

8

MICHAEL F. C. BOURDILLON

needed to be improved. On the other hand, they were in these schools by their own choice and would have been worse off in the long term without them (Bourdillon, 2000b). Even when work interferes with development and education in particular, several studies have pointed to the harm that children sometimes suffer from when they are thrown out of work (see Bachman, 2000; Bourdillon, 2000a, pp. 12–15; Save the Children, 2000, pp. 40–42). It is not clear that a legal minimum age for employment always works in the best interests of children. For children who need an income, child unemployment rather than child labor is the major problem (Andvig, 1998, p. 327). The term “child labor,” however, still evokes images of children being exploited and abused in situations that impede their mental and social development. There have consequently been attempts to distinguish “child work,” which is benign, from “child labor,” which is harmful work. Such a distinction enables us to maintain a program for the elimination of child labor while acknowledging that work can sometimes be beneficial for children. This distinction, however, focuses on the label that is applied, rather than on the contexts of the children and the conditions of their work. It diverts attention from what precisely is to be eliminated, which is not the work or labor, so much as the harmful conditions under which it is sometimes performed. Further, in other contexts (and especially legal contexts), “labor” refers to formal employment, and the distinction between “child labor” and “child work” easily confuses harmful work with formal employment. I follow the guidance of the Save the Children Alliance (2003) and avoid the term “child labor.” I use the term “work” to cover all the work that children do, in the home and out of it, paid or unpaid. I leave the judgment of whether it is good or bad for the children to be determined by the contexts in which the children find themselves. If the available alternatives to working leave the children with a worse life and fewer chances of developing, then the work is beneficial to them. If there are better alternatives, these should be made more available so that more children can avoid work that disrupts their development. In some cases, the work is unequivocally harmful (as is prostitution in the context of an epidemic of HIV/AIDS), and the need to find alternatives is desperate. One important aspect of the context of child work is the way the children themselves perceive it. The United Nations Convention on the Rights of the Child states that children’s opinions should be given serious consideration in matters that affect their lives (Article 12). There has been some debate on whether children have the right to choose to work. Children’s opinions are not consistent and can be manipulated by supporting adults.1 The opinions of children, like those of adults, can be colored by cultural assumptions. We need to listen to children, but also to direct them.

Working Children in Zimbabwe

9

Since much child work is hidden, sometimes not even acknowledged, it is difficult to obtain accurate statistics on the scale and extent of child work. According to a survey in September 1999, 26.3% of Zimbabwean children aged 5–17 were engaged in “economic activities,” with just over half of these working for over three hours a day (Government of Zimbabwe, 1999, p. xiii). Housework was excluded and the survey was conducted at a time when there was little work on rain-fed family agriculture. These figures do not therefore fully reflect the contribution of children’s work; they nevertheless indicate the widespread involvement of children in work, and suggest that the case-study material in this chapter represents the situations of a large number of children. Until recently, Zimbabwe has been among the more wealthy countries of subSaharan Africa, in its high rates of children attending school (79% reaching grade 5) and its relatively high GNP per capita ($610, ranking fifth in Sub-Saharan Africa). The percentage of children aged 10–14 in the labor force was recorded as 27% in 1999, below the average of 30% for the sub-Saharan region.2 While the situation may have been better than in other countries, there have always been many poor people in the country and many children working. Now, the collapsing economy in Zimbabwe is bringing increasing levels of poverty, and poverty in countries relates to levels of child labor. The collapsing economy leaves many adults without formal employment in a situation of high inflation and decreasing resources for social welfare. In a situation like this, children, like adults, have to find ways of coping with minimal outside support, whether they are on their own or have to contribute to the livelihood of an impoverished family. The protected childhood that is the ideal in more affluent societies is not available to most children of Zimbabwe. A second factor that is contributing to growing levels of child work is the HIV/AIDS epidemic that has been decimating the adult population: around 20% of children in Zimbabwe have lost at least one parent. In the late 1990s, together with several students and colleagues, I collected material on working children in rural areas in the southeast of the country, and in some urban areas (see Bourdillon, 2000a, b). We chose a variety of situations in which children were working, and in most cases spent time living in the communities concerned and getting to know the working children and their families. Recently, I have been working with the Zimbabwe Office of Save the Children Norway, helping trades unions to establish clubs for working children in agriculture and domestic work in several rural and urban areas, and collecting information on some of the children. While I do not have a systematic representative sample of working children in Zimbabwe, data come from a wide range of situations and enable me to place the children’s work in the contexts of their lives and their families. I pay attention to the children’s perceptions of their work, often as necessary and normal, and beneficial to themselves and their families.

10

MICHAEL F. C. BOURDILLON

I point to the difficulty of drawing a line between what benefits the children and what harms them.

CHILDREN IN THEIR HOMES In subsistence economies throughout the ages, children have contributed their labor to the family livelihood insofar as they have been able. Children as young as five start helping with household chores. In agricultural societies, children help in the fields as soon as they are able, planting, weeding, reaping and protecting crops from birds and other pests. Children are involved in herding domestic animals at a young age. Young children care for even younger ones, freeing older people for productive work. This leaves some children, and particularly girls, with little free time (Reynolds, 1991, pp. 41–92). Where families have specific skills, such as music or healing or some craft, children start to learn the skills and participate in the work (Reynolds, 1996, p. 38). This pattern continues in contemporary countries where the bulk of the population remains in agriculture. In poor families, such work can be substantial. An example is Catherine’s family in a remote rural area in the east of the country. Catherine’s father had three wives and seventeen children, and lived off his small piece of land (Mangoma & Bourdillon, 2002, pp. 18–21). Catherine was twelve years old and the eldest daughter of the second wife. She performed chores of cooking, cleaning the house and utensils, fetching firewood and water (the firewood was occasionally for sale), all of which amounted to just under 40 hours a week. She also sometimes undertook contract agricultural employment to obtain money for her school fees. She complained of getting tired and having nose bleeds in the hot weather, but commented: My work is important in the household because I help my parents with most of the work, making it easier for them . . .

In spite of all this work, she was placed fifth in a class of roughly 50 children. Her elder half-brother spent less time on housework and more time in the fields, amounting to over 30 hours a week altogether. He commented: I play an important role within my family because my parents depend on me for doing some of the household chores.

Even six-year-old children in this family were given chores to do. The work of a child in maintaining a family can be a source of pride and status within the family. Nhamo, also living in the rural east of the country, exemplifies this point. He was 13, the oldest child in his family, and still at primary school.

Working Children in Zimbabwe

11

His father was dead, and he and his siblings lived with his elderly grandparents, together with others of their grandchildren. Apart from doing most of the work in the homestead, he was the principle breadwinner for the family, contracting in the surrounding community various types of work, particularly work in the fields of small-scale commercial farmers. The work interfered with his schooling, which was also interrupted at times by lack of funds. But he was held in esteem by his grandparents and by the other children of the household (Chirwa & Bourdillon, 2000, pp. 137–139). The case of Laina illustrates how household work can become particularly strenuous in the absence of a parent. Her mother was sick in hospital, and Laina, who was 16 and the eldest girl in the family, had to run the household, which occupied over 40 hours a week (Mangoma & Bourdillon, 2002, pp. 24–26). I will be exhausted from the work so that I can hardly read, for me I have to be like a mother when my mother is sick.

She also helped in taking care of her grandmother, who was too old to do any work. She acknowledged that the work at home interfered with her schoolwork, but pointed out that she had to work to ensure the survival of the household. Tinashe is another child who was compelled to take on a role normally ascribed to adults. He was thirteen and lived with his father’s elder brother, since both his parents were dead. He worked for the household, both in the fields and in the home. The fields produced some food, but not enough cash income for the family’s needs. They received occasional help in food from the Department of Social Welfare, but not enough to live on. So Tinashe also engaged in contract work to keep the family going and to raise his school expenses. He did not, however, expect to be able to complete his schooling. My life is different from most children . . . because these children have a mother and a father, so it makes all the difference in the world. I also have a big responsibility because my uncle and grandfather are always sick and so I have to help them by cooking for them, fetching water for them to bath, washing their clothes, plowing, planting and harvesting our small plot. All these things I have to do. I cannot refuse.

Many children, and particularly girls, are heavily involved in caring for sick adults (Robson, 2000). Sometimes, the child is taken out of school to care for a parent or relative. In the context of the HIV/AIDS epidemic, it is not uncommon for an adult to need full-time care for an extended period of time. Both adults and children take for granted the responsibility of children to look after a sick relative when there is no one else to do this. While such work can interfere with social life, and particularly with schooling, children do not complain and sometimes comment that that the work is rewarding.

12

MICHAEL F. C. BOURDILLON

No one questions that it is appropriate for a child to care for a sick parent, although we would prefer other support to be available, giving the child freedom to socialize with peers and to continue with their schooling. The cultural expectation that children must support their kin goes further. Takura was 16 and spent around 35 hours per week on work at home besides his schoolwork (Mangoma & Bourdillon, 2002, pp. 26–28). He helped in household chores like sweeping, cooking and washing clothes. He worked in the fields and fetched firewood and water, for his own household and that of his grandmother. He also ran errands for his father’s brothers: he calls them “father” (as is normal in the cultures of Zimbabwe) and would not think of disobeying. His mother explained that her children are also children to their relatives and therefore under the authority of senior relatives. His father commented: Children work for their relatives because they are our kin and they [the children] cannot say “no” . . . it is part of our culture.

While Takura did not question his obligations to senior kin, perceptions of children and adults on the cultural responsibility to support kin do not always coincide. Buhle was a girl sent by her parents to look after an aunt in another city (Rurevo & Bourdillon, 2003, p. 33). The aunt had no one else to help in the home, and was often away on trading business. When she asked for help, it was hard for Buhle’s parents to refuse. Buhle was supposed to be going to night school. In the event she was not paid and was kept too busy for school, and so decided it would be better simply to support herself on the streets. In other cases, children can be made to work for kin who appear to be looking after them. An example was Justin, who was 17 when I met him. His father had died, and his mother had re-married and was living elsewhere. His father’s family refused to support him and he had been staying with his maternal grandparents. When his grandfather died, the heir refused to support Justin and demanded that he earn his keep by working several days a week unpaid on the family fields. He dropped out of school and tried to earn some money of his own in contract work when he had time off from his cousin’s fields.3 I have shown in this section how work at home is taken for granted by both children and adults. When the family is in trouble through death, sickness or simply poverty, such work can interfere with formal schooling and other activities of childhood. But the children still accept their work as necessary, and it can become a source of self-esteem and esteem in the eyes of others. In certain situations, the children see themselves as being exploited by their kin. For legal purposes, such exploitation still appears as work in the home, and is easily missed in campaigns against child labor. Indeed, the line between what is legitimately demanded of children and the exploitation of child labor in the home remains blurred and is difficult to define.

Working Children in Zimbabwe

13

CHILDREN EARNING MONEY We have been looking at work in the home, which is largely domestic work, but which includes some economic work on family fields and gardens that contributes directly to the family livelihood. I have mentioned children doing occasional contract work to supplement family incomes. The help of children in their homes moves easily into helping parents and other adults to earn cash incomes. Betty was 13 when we met her (Chirwa & Bourdillon, 2000, pp. 131–133). The family income came largely from hiring out the family oxen to members of the local community. While their father controlled all the financial aspects of this work and looked after the cattle, it was the role of Betty and her brother to lead the oxen in their tasks. This is considered work for children rather than adults. Such work is ideally undertaken in the morning, before the oxen are tired by the heat of the day. This meant that Betty and her brother often had to rise early, to do a couple (or more) hours of strenuous work before going to school in the morning, leaving them tired and sometimes inattentive during class. Betty commented: Working is not enjoyable, but for me I have to work because if I do not, all the things I have now will not be available. I will not be able to go to school, have uniform and food. Waking up very early in the morning to work before we go to school is not very good but there is no choice at times.

In the urban context, there is no opportunity to contribute to the family livelihood by working on the family gardens and fields. Instead, children in poor families may help in their parents’ activities to generate cash incomes. Informal trading can be combined with schoolwork, and many children point out that it can be combined with play. Market places often attract children, and vending gives them a reason to be there. In some cases, the work is serious and takes up much of the child’s time. Wadza, for example, was 13 and helped her mother who was trading in firewood in a Harare suburb (Mapedzahama & Bourdillon, 2000, p. 32f.). She came to the stall after school at midday, and often stayed until seven in the evening, although she was able to play with friends at the same time. She complained of verbal abuse from customers, and of drunken men urinating near where she had her wood. She also acknowledged that the work interfered with her schoolwork. Nevertheless, she would continue to help her mother because this was the only source of family livelihood. When women take up contract employment to earn cash incomes in rural areas, they are frequently accompanied by young children, who may spend time in idle play or often help their parents in the contract work. Growers of tea and coffee normally reap their crops by paying workers on the weight of the crop that they pick. Each contract worker is assigned a card or “ticket” on which their pickings are recorded and on the basis of which their wages will be calculated. It is very

14

MICHAEL F. C. BOURDILLON

easy for children to add to their mother’s pickings, which allows them to contribute to family income without the estate acquiring a stigma of employing child labor. As the children grow older and pick more, they often want to acquire recognition for the work they do, as well as a degree of control over their earnings, by being registered on their own personal tickets. Both children and managers regard it as unfair that this is not allowed if the children are below 15, the legal minimum age of employment. Some children point out that learning a trade with their parents is more useful than schoolwork. Vee was helping her mother in informal trading in a Harare suburb, and acknowledged that the work was interfering with her performance at school. Nevertheless, she was learning how to earn an income and said that this was more important than what children are taught at school (Mapedzahama & Bourdillon, 2000, p. 38). Thomas was taken out of school by his father after completing primary level, to work at a sewing machine in his father’s business in a small town (Mutisi & Bourdillon, 2000, p. 91f.). He pointed out that he was learning a skill that was earning an income, and he was the envy of his peers who were still at school. He would like to learn more about clothes design and improve his tailoring skills, but he had no desire to go back to school.

ON THEIR OWN Some children work on their own on the streets and constructively contribute to the family livelihood. Catherine was running a vegetable stall in Harare at the age of twelve. With her mother and elder sister, she had fled an abusive father, and come to Harare, where her mother earned a living running the stall. Then her mother fell sick and the family livelihood was threatened. In order to allow her sister to complete her public examinations in secondary school Catherine dropped out of school to run the stall (Rurevo & Bourdillon, 2003, pp. 43, 44). My sister is about to finish school, and if she gets a better job she will be able to send me to school and look after our mother. I have to look after her now and I await the same from her. Moreover, I want to prove my father wrong. I want to prove that girls are as good and responsible as boys are.

She was able to go home to her mother at night, and her sister sometimes helped with the stall. Her mother’s friends in the market place also supported her and protected her from anyone who might try to take advantage of her. Catherine commented: These women are like mother hens to me . . . If it were not for them, my mother would have died and my sister would have been out of school. They contributed much towards my mother’s

Working Children in Zimbabwe

15

medication. Sometimes I wander and play in town, only to come back to find that they have sold some of my wares on my behalf in my absence.

Another example was Munya, who was 17, and had left school two years previously when his mother died and the family had no regular income (Mapedzahama & Bourdillon, 2000, pp. 35–37). He lived with his unemployed father and 11-yearold brother. He started earning money from washing cars at an urban centre, and then traded in snacks of roasted pieces of chicken. He did not enjoy his work and would like to go back to school (though not in town where he would be ridiculed for being old for his class). Again his income improved the living standard of his family and enabled his brother to continue his schooling. Orphans in particular often have to support themselves with trading, even when they have formal guardians to provide some support. Anna started trade at a rural center at the age of seven, when her father died (Mutisi & Bourdillon, 2000, pp. 83–85). Six years later she was still vending, to help the household, and still in school. Sometimes she had to miss school when there was urgent need in the family for cash, but for the most part she managed to combine school and work. In urban areas, such vendors are subject to harassment from police and other officials, who say they should not be on the streets. The children acknowledge that work is interfering with their education and limiting possibilities for their future. But when earning on the streets is the only source of livelihood for them, they resent harassment from authorities: We are just simple vendors. We do no one any wrong. Why harass and arrest us? (Muzvidziwa, 2000, p. 73)

Not all sources of income on the streets are benign. Without support from family and friends, girls on the street invariably have to rely at some time on boys and men, and almost all get drawn into prostitution (Rurevo & Bourdillon, 2003, p. 55). The girls do not like talking about this aspect of their lives and resent the way people look down on them. But they have no alternative. As one girl put it: My sister, life in the streets is tough, one has to devise several strategies to survive, but it is heavy (ibid, p. 31).

Another commented: Sister, since I am a girl living on the streets with no other means of generating income, how do you think I earn that income or my living? (ibid, p. 27).

Boys on the street have greater opportunities for earning money. Apart from easier access to trading (the boys sometimes hound girls off favored places for trading), they earn money from parking and guarding cars in the city center. Nevertheless, they also sometimes prostitute themselves for a good meal and a comfortable place

16

MICHAEL F. C. BOURDILLON

to wash and sleep for the night. Or perhaps children depend on adults, such as security guards, for a place to sleep and temporary security, and these adults demand in return that the children (boys as well as girls) give themselves to paying clients. Similarly, older children might take charge of younger children to obtain an income in this way. The security that children obtain from their “guardian” is very short term, in a society where around 30% of sexually active people have the AIDS virus. While it is not tolerable to leave children in such a situation, whatever their own wishes, it is not clear what can be done for them in the absence of effective policing and in the absence of adequate welfare provisions.

EMPLOYMENT Some child vendors are working on behalf of adults, who purchase the goods to be sold and who often take the major part of the takings. The children are effectively employed by the adults, sometimes for a small wage, but more usually for a portion of the profit. In either case, income from such employment is very low. Incomes in the informal sector are generally low, and in a weak economy people survive by unregulated work requiring long hours for little pay. For children who need incomes, the lack of regulation is an advantage, since they can find employment in the informal sector when formal and regulated employment is made impossible by the law. In this way, we find children working in dangerous conditions for low pay in informal mining (McIver, 2000). At a workshop on working children in Harare, children employed in informal mining pointed out that they had no option but to accept such employment, since they and their families needed money to eat. They complained about the conditions of their work and asked to be allowed to work in the formal mining sector where working conditions are better, safety precautions are more thorough, and wages are higher. This is impossible since to the outside world underground mining is one of the worst forms of child labor, to be eradicated immediately. To the children, it would be better than the alternatives available to them. Another area in which many children find unregulated employment is in domestic work. Children from poor families stay in the home of a family that is better off, and earn their keep by performing chores. In the rural areas, such chores include agricultural work on family land. In urban areas, it involves mainly household work and caring for small children. The employers may be kin of the working children, particularly when the children have been orphaned. Alternatively, the employer may be an acquaintance or someone contacted by the child or parents through a third party. The practice fits into a strategy by poor families of placing children with others for greater security, and for support

Working Children in Zimbabwe

17

that the parents cannot always provide (Van der Waal, 1996, pp. 43–45). The relationship between children and employers is often expressed in fictive kinship terms. Sometimes children work simply for their keep. More frequently wages are paid besides the keep either to the children or to their parents or guardians. Such arrangements combine employment with patronage. Employers speak of the children becoming members of the family, and emphasize the help they provide to children in need. In a survey of child domestic workers,4 roughly half of the children said that they were happy in their place of work, and many commented that they were treated like their employers’ own children. They spoke well of their employers, and the patronage can work well for some children. Particularly those in their mid-teens and older are likely to be in employment of their own choice: they do not want the employment to be terminated and to be forced back to an impoverished home. When the children were asked if they would like to stop working, a frequent response was, “What would I do?” Some who were unhappy with their situation would like to leave if they could find better employment. There is, nevertheless, often an element of duress in the choice to seek employment. Children said they did not want to stop working because it gave them food, or money for clothes. A number pointed out that if they left their employment, they would have nowhere to live. Mary, for example, was 16 and working in a house in a poor Harare suburb. She found nothing that she liked in her work. She did not like going to bed while she was still hungry, getting up while it was still dark, having to sleep with only one blanket in winter, having no jersey to wear, not being treated like the others. But she did not want to leave her job: because if I stopped work, I would have nowhere to go.

She explained, My parents [in a remote rural area] do not have enough money to send me to school. They do not find enough food.

Another example is a 17-year-old girl, who has no father, and does not like her work because both her employers (husband and wife) made her work hard and scolded her often. But she could not leave her job: If I stop working, there is nowhere for me to live, and I will not find money for the things that help me.

Child domestic work is unregulated and open to abuse. Sometimes, the children have no say in the arrangement made by their parents or guardians. Formal contracts of employment are rare, and most children are on call to do chores at any time of the day or night. Usually they do not have specific days or times when they are off

18

MICHAEL F. C. BOURDILLON

duty. Often their sleeping arrangements, and even the food they eat, are inferior to those of the children of the household. They are frequently subject to various kinds of abuse, including insulting scolding and even corporal punishment when they make mistakes. Children complain of the way in which employers, and other family members, treat them with contempt. Some children nevertheless take this kind of treatment for granted, since they see no alternative to their present situation. It is only when they meet with other children in similar situations that they are able to compare their treatment by different employers, and begin to assess what is normal and acceptable treatment. The majority of child domestic workers are out of school, and many complain that they would prefer to continue their schooling if they had the means. In our sample, only a quarter were still in school. Some of these had their school expenses covered by their employers and several pointed out that if they stopped working they would not have money for school. A 17-year-old explained that he could not leave work because that is where the money comes from for his younger brother’s schooling. Formal employment can be combined with education in other ways. Certain estates run schools at which the children register on condition that they work for the estate on a contract basis (Bourdillon, 2000b, pp. 159–167). The children attended school, with better facilities than other local schools, and earned all their school expenses with a considerable surplus besides. Sometimes this surplus was used to help families, and the education of younger siblings in particular. The estates, on the other hand, had a stable supply of contract labor for the picking season, which compensated for the trouble and expense they incurred in running and subsidizing the schools. Local people acknowledged that life at these schools was hard, and such a system is open to abuse. Nevertheless, the children were almost invariably there at their own choice, and the local communities regarded the schools as a benefit. Sadly, when the minimum age of employment was raised to 15, many children who were dependent on these schools for their secondary education had to wait some years after completing their primary schooling.

CONCLUSION The African Charter on the Rights of the Child, approved by the Organization of African Unity in 1990, pointed to the responsibility of children towards their families and communities (Article 31). This reflects a widespread view that children have a contribution to make towards their communities, and that rights should be seen in conjunction with obligations.

Working Children in Zimbabwe

19

A comparison of perceptions about child labor of school children in Zimbabwe and Norway, found that compared to the home duties of Zimbabwean children, those of Norwegian children were more clearly defined, were frequently the result of negotiation, and were commonly paid for (Nilsen, 2002, p. 68). Whereas Norwegian society pays attention to the individual rights of children, Zimbabwean society pays more attention to the family context that gives the children their identity and their life. The Zimbabwean children took for granted their contributions towards their homes, and did not mention such work in the contexts of child labor. While this sense of children’s responsibility can result in children being better integrated into their communities, it can also result in particular children suffering from loss of social and educational possibilities. In Zimbabwe, as in many societies, people take for granted that children should work. People often complain that the campaign against child labor is an attempt by Western nations to impose their values on other peoples, without understanding the cultures and the contexts of the children affected by the campaign. Any attempt to improve the situation of the children must respect the cultural values of these children. While children and adults acknowledge that work sometimes interferes with education, they point out that through this work children often earn the money that makes education possible. On the other hand, cultural values often support the interests of certain people to the detriment of others. Much has been written about how ruling classes maintain dominance over workers, or males dominate women, through a variety of cultural ideologies. Similarly, cultural values sometimes allow adults to exploit subordinate children. Cultural practices are presented to children to be accepted without question: children who reject such practices are depicted as rebels and traitors. Under such influence, children learn to live with exploitation and abuse. Although the children may learn to tolerate exploitation by adults, there is no excuse for adult society to tolerate such practices. But we should be careful to focus our condemnation on the exploitation and abuse of the children. Their work should be valued and rewarded. Their responsible contributions to their own livelihood and that of their families should be appreciated. Cultural change results when younger generations selectively maintain and reject different aspects of the culture passed to them by their elders. When politics, the economy, and disease disempower adults, children have a particularly important role in shaping the society they are to live in. In such a situation, the best hope of helping children in difficult circumstances is to build on their resilience, and to support them in their efforts to provide for their own livelihoods and their development (Bourdillon, 2004; Fraser & Gakinsky, 1997). This involves helping them to see that some things can be changed and that they can as a group influence events in their own lives.

20

MICHAEL F. C. BOURDILLON

If we are to guide children in discerning what is good in their lives and what needs to be changed, we need to be careful and accurate in our own thinking. The long-term prospects of the next generation will not be improved by the elimination of work or labor from their lives. It is not helpful to eliminate work that benefits their livelihood, nor to eliminate the recognition of such work through payment in employment. We need to eliminate physical danger in children’s work. We need to eliminate long hours of work that hinder their development. We need to ensure that children have time, energy, and facilities to learn. They need time to rest and time to be creative. We need to eliminate bondage. Children need opportunities to make choices, and to have say in the outcome and the betterment of their childhood.

NOTES 1. See the different statements by South American children on child labor in Post (2002, pp. 45–56). 2. The figures are for 1999, coming from World Development Indicators in World Bank (2000). See Tables 1, 3, 5, and 6. 3. For other examples of work masked as kinship relations, see Porter (1996) on Kenya, and Bass (1996) on Senegal. 4. The Zimbabwe Domestic and Allied Workers’ Union, supported by the Zimbabwe office of Save the Children, Norway, is currently encouraging the formation of clubs for child domestic workers. Of these children in clubs, 116 have completed a questionnaire on their work, but the results are not yet published.

REFERENCES Andvig, J. C. (1998). Child labour in Sub-Saharan Africa – An exploration. Forum for Development Studies, 2, 327–362. Bachman, S. (2000). A new economics of child labor: Searching for answers behind the headlines. Journal of International Affairs, 53(2), 545–572. Bass, L. E. (1996). Beyond homework: Children’s incorporation into market-based work in urban areas of Senegal. Anthropology of Work Review, 17(1–2), 19–25. Bourdillon, M. (2000a). Introduction. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 1–24). Harare: Weaver. Bourdillon, M. (2000b). Children at work on tea and coffee estates. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 147–172). Harare: Weaver. Bourdillon, M. (2004). Children in development. Progress in Development Studies, 4(2), 99–103. Chirwa, Y., & Bourdillon, M. (2000). Small-scale commercial farming: Working children in Nyanyadzi irrigation scheme. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 127–146). Harare: Weaver.

Working Children in Zimbabwe

21

Fraser, M. W., & Gakinsky, M. J. (1997). Toward a resilience-based model of practice. In: L. D. Kirby & M. W. Fraser (Eds), Risk and Resilience in Childhood: An Ecological Perspective (pp. 265–275). Washington: National Association of Social Workers. Government of Zimbabwe (1999). Zimbabwe 1998: National child labour survey country report. Harare: Central Statistics Office. ILO (2002a). A future without child labour: Global report. Geneva: ILO. ILO (2002b). IPEC action against child labour: Highlights 2002. Geneva: ILO. McIver, C. (2000). Child labour in informal mines in Zimbabwe. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 173–186). Harare: Weaver. Mangoma, J., & Bourdillon, M. (2002). The work of children in impoverished families. In: P. Hebinck & M. Bourdillon (Eds), Rural Livelihoods in South-eastern Zimbabwe (pp. 13–35). Harare: Weaver Press. Mapedzahama, V., & Bourdillon, M. (2000). Street workers in a Harare suburb. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 45–58). Harare: Weaver. Mutisi, M., & Bourdillon, M. (2000). Child vendors at a rural growth point. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 75–94). Harare: Weaver. Muzvidziwa, V. (2000). Child vendors in the streets of Masvingo. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 59–74). Harare: Weaver. Nilsen, A. C. E. (2002). Negotiating children’s work: A comparative study of children’s work in Norway and Zimbabwe. Post-graduate thesis, Department of Sociology, University of Bergen. Porter, K. A. (1996). The agency of children, work, and social change in the south Pare Mountains. Tanzania. Anthropology of Work Review, 17(1–2), 8–19. Post, D. (2002). Children’s work, schooling, and welfare in Latin America. Boulder: Westview. Reynolds, P. (1991). Dance civet cat: Child labour in the Zambezi Valley. London: Zed Books. Reynolds, P. (1996). Traditional healers and childhood in Zimbabwe. Ohio: Ohio University Press. Robson, E. (2000). Invisible careers: Young people in Zimbabwe’s home-based health care. In: M. Bourdillon (Ed.), Earning a Life: Working Children in Zimbabwe (pp. 109–126). Harare: Weaver. Rurevo, R., & Bourdillon, M. (2003). Girls on the street. Harare: Weaver. Save the Children (2000). Big business, small hands: Responsible approaches to child labour. London: Save the Children. Save the Children (2003). Children and work: Save the children’s position on children and work. London: Save the Children. Van der Waal, C. S. (1996). Rural children and residential instability in the northern province of South Africa. Social Dynamics, 22(1), 31–53. World Bank (2000). Entering the 21st century: World development report 1999/2000. London, etc: Oxford University Press.

TOYS AND GAMES: CHILDHOOD IN ˜ THE PARQUE DAS NAC¸OES FAVELA IN BRAZIL夽 Ethel Volfzon Kosminsky and Laura Daniel INTRODUCTION When observing the work of Brazilian Sociology and Anthropology, it is apparent that the study of childhood, particularly concerning play and toys, has received little attention. A quick evaluation in this area shows that few existing works appear in the fields of Psychology and Education, the first mostly concerned with the psychology of development and the latter with the question of learning.1 The first sociological research in the area of childhood began in the 1970s and covered such fields as child labor, institutionalized children, homeless children, the child’s role within the family and his relationship with school, violence against children, child pornography, and juvenile delinquency.2 Anthropological research started in the 1980s and covered such subjects as native children, African-Brazilian children, homeless children and children who live in favelas.3 In the last decade, the International Labour Office (ILO), affiliated with the United Nations, has completed research about child labor and the problems it inflicts on children. This effort by the ILO has the support of politicians and its coverage in the local press has increased the desire for more research, primarily in the university.4 夽 An

earlier version of this paper was presented at the ISA conference, Brisbane, Australia, July 2002.

Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 23–41 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10003-2

23

24

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

This concern with poverty and the oppressed, connected to the increasing gap between social classes has reached all of Latin American Sociology, not only Brazilian. As Alejandro Portes (2002) has pointed out, Latin American Sociology has found itself overwhelmed by social problems caused by poverty, social inequality, and violence. In addition, it has had to deal with financial problems that have placed its research in jeopardy; government financial support is always precarious and often insufficient. Moreover, for decades it has also confronted political impediments from the authoritarian governments of both the left and the right. For all of these reasons, Latin American sociology, in particular Brazilian, has dealt with questions of marginality. In doing so, according to Pereira de Queiroz (1992), they have narrowed the themes to be investigated and have minimized the study of other social groups as if there were no reason to conduct further investigations into these other groups because they thought their conditions were already so well known. The influence of orthodox Marxism on Brazilian and Latin American Sociology, on the other hand, led to the creation of a large interpretative framework that has retarded scientific advancement. According to Portes (2002, pp. 6, 7) there are three main reasons behind this trend: First, the substitution of a theoretical/deductive logic for an empirical/inductive one. Second, the deproblematization of the world. Since social phenomena can be readily explained deductively, there is little need for empirical study. Third, the reification of concepts. Because of their globalizing character, grand theory often acquires a life of its own where concepts become isomorphic with reality itself.

In addition, the great influence of Marxist theory, in which the workers are responsible for social change, devalues the importance of other social participants such as children, women, youth, and the elderly, who are not considered worthy of social investigation. Even in the case of youths, studies have concentrated on the theme of juvenile delinquency.5 Youths only appeared in focus if they were revolutionaries and followed the ideology of the student movements of the 1960s. Otherwise, they were studied as if they were a group alienated from revolutionary ideology.6 Recently, sociologists and anthropologists rejected this point of view showing that it became frozen as an ideal model of youth behavior, and have focused on new topics such as youth and funk and hip hop, and youth and violence, in this last case as agent of violent acts and, on the other hand, as a target of adult violence.7 In most of this research, the child is seen as a victim of society: victimized by family violence, by the police, by adults in general, and finally as passive recipients of adult actions. In the field of History there has been growing research into the family and children, whether abandoned or interned in institutions.8

Toys and Games

25

In the 1940s, when the boundaries between Folklore and Sociology were not so clear, Florestan Fernandes (1979), a leading Brazilian Sociologist, completed research into children’s games on the streets of Bom Retiro, a S˜ao Paulo immigrant neighborhood inhabited by Eastern European Jews, Italians, Japanese, and AfricanBrazilians, all members of the working class or of the lower-middle class. Fernandes observed the socialization process of children and studied the Iberian origin of children’s play songs. In contrast to studies of that period that treated the socialization of children as if they were mere recipients, Fernandes showed the child to be a social actor, transmitting the social culture of Portuguese-Brazilian society to his or her immigrant parents. Although, he did not develop this idea in great depth, his research represented a methodological innovation, according to his Professor Roger Bastide (1979, p. 154): To study children, it is necessary to become one. I mean that it is not enough simply to observe children from a distance, just as it is not enough to accept their invitation to play with them. It is necessary to penetrate into that magic circle that separates us from them. In order to feel their worries, their passions, we must have fun with them. Not everyone is capable of doing this. The most important attribute of Fernandes’s work is that it is the result of a profound investigation. The author entered into a great conspiracy with these children.

Works about children’s folklore have been rare and hard to find. Sometimes, they have been limited to chapters of books that cover folklore in general.9 Research about toys and children’s games can be conducted using children’s literature or autobiographies that refer to childhood.10 Children’s literature and those books recounting childhood memories most often refer to rural games with dolls made from pieces of fabric and corncobs. In addition, children listened to stories told by grandparents and old black housemaids, told in a mixture of Portuguese, African, and indigenous legends. We can presume that throughout the centuries children have played on the streets of small cities and in the backyards of their houses with neighborhood children. Certainly, this must have been true until the 1950s and 1960s when little girls sang old songs and played on the streets of large Brazilian cities. Young boys also played games with girls; older boys likely played soccer together on the street. Boys and girls played games of hide-and-seek. Young girls played with their dolls on the street and played house at each other’s homes. Much like the homemade dolls, the boys made their own balls out of old socks. Some dolls were store bought and had heads made of porcelain; others were made of cheap plastic. Rapid urbanization, accompanied by the vertical growth of cities and their increased density, chaotic vehicular traffic, and the growth of urban violence led to the increasing isolation of middle- and upper-class children, who currently play in

26

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

playgrounds within the gated communities where they live or in playgrounds found in private clubs or shopping malls. Only children living in poor neighborhoods currently play on the urban streets. Yet, what is a toy? Cascudo (1984, p. 53), the Brazilian folklore scholar said: A toy is as activity and it is an object, a car, a monkey on a string, a doll with a music Box, a toy car, a witch, a ball, a wooden horse. To play is to run, sing, pull a wooden elephant attached by string or use a wooden stick as a horse, while pretending to be a brave warrior.

In discussing the arguments raised by Brian Sutton-Smith, Mergen (1995, p. 257) has elaborated that children’s games have always been defined using the term “rhetoric of progress, the idea that play prepares the child for adulthood.” He also points out that: “children’s play is more than a rehearsal for the drama of adult play, that it may also involve the resolutions of conflicts, the building of community, and pure creativity.” Thus, this is the sense of play as a child’s game understanding social interaction permeated by symbols, beliefs, attitudes, fantasies and, above all, by fun, enjoyed by the children that we followed in this article. This work researches the social relations of children between the ages of seven and ten who live in the favela called Parque das Nac¸o˜ es. It attempts to interpret how children play with their toys despite the harsh conditions of their lives. These children have established a series of relationships, including social relationships with their families and infant groups, in a way special to that of children who play and create games. Moreover, they have developed gender relations while playing games. We also study whether mothers have transmitted their own games to their children and whether the games that children play demonstrate violence. Above all, we have talked with the children and observed their games and interviewed mothers, fathers, stepfathers, grandparents, and older brothers and sisters who live in the same favela.

THE CITY OF MARILIA AND ˜ THE PARQUE DAS NAC ¸ OES FAVELA The city of Marilia, where we completed this research, in common with nearly all the cities of Brazil, shows the inherent contradictions in Brazilian society and the abyss that exists between social classes. The Federal Census of Population completed by IBGE (Brazilian Institute of Geography and Statistics) in the year 2000 counted 198,658 inhabitants in Marilia, which can be divided into neighborhoods of the upper class, middle class, workers settlements that were constructed by the mayor’s office, and favelas. In 1999, the IBGE ranked Marilia

Toys and Games

27

as the most economically developed city in the entire state of S˜ao Paulo because it contains diverse industries in food production as well as three universities, which attract a large, floating population of young adults that greatly aids commercial activity. According to surveys conducted in the favelas by the Marilia Department of Social Welfare there were eighteen favelas within the city in 1994, nineteen in 1997, and twenty according to the most recent tally. According to the data compiled in 1999, the Parque das Nac¸o˜ es favela is the second largest in the city, containing 497 people or nearly 10% of the total of 4,907 people who live in favelas throughout the city. An adjoining neighborhood in the city, which also goes by the name Parque das Nac¸o˜ es, contains a complete infrastructure, including an electric energy network, a system of water supply and sewage disposal, paved streets and so on. It maintains a public school and two public health offices, including one aimed solely for the health of mothers. There are also supermarkets, stores, and other commercial establishments. In contrast, the favela Parque das Nac¸o˜ es only provides electricity and systems of water supply and sewage disposal. It has no paved streets and collects garbage only one day per week. Moreover, there are no commercial establishments in the favela besides two bars. The majority of families who live in the favela Parque das Nac¸o˜ es consist of migrants from the Northeast of Brazil, people who have come from neighboring cities, or those who have arrived from other periphery settlements and the most needy sections of the city. In all the interviews collected, we determined that some mothers in the favelas worked in industrial plants, but the majority worked at home or else possessed some link to obtain sporadic informal employment as a per diem housemaid. The fathers either worked within the industrial district’s food production plants or in home construction as laborers, house painters, plumbers, or electricians. The families were composed generally of a mother, father or stepfather, children, grandmothers and, in some cases, a mother’s brothers. The grandmothers had an important role in the favela, for they were generally retired and could watch after their grandchildren while their own daughters worked.

CHILDREN AND THEIR FAMILIES ˜ IN THE PARQUE DAS NAC ¸ OES FAVELA The researchers made the first contact with the residents of the favela in 1999 and all the families, adults as well as children, were always very friendly and receptive to the project. The first step in conducting the interview was to visit all the houses of the favela and search for children between the ages of seven and ten. We found eighty-eight such children in fifty-two homes, and selected twenty-six houses to

28

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

be visited the first six months of the year and twenty-six during the second half of the year. All the children of this age group were limited to those who attended public school in the afternoon. As a result, we completed interviews with families during the morning and also on Saturdays. We formulated some questions, but were not strictly bound to them. Although we utilized a tape-recorder, we also let the respondents speak freely. This last technique permitted us to learn other information during the course of the interview and raised the possibility that we could collect not only explicit material, but also the implicit, the subjective and many times, as Pereira de Queiroz wrote (1991, p. 23) “the collective unconscious and the archetypal.” Completing research with children implies certain specific problems, such as how to maintain their interest when they became distracted in the middle of a conversation by a sudden noise or movement and would lose their concentration. According to Kosminsky (1992), the tape-recorder facilitates access to the child because s/he can enjoy the satisfaction of hearing her/his own voice being taped. The idea was to establish an informal, friendly meshing of interviewer and interviewee and remove the formality of direct questions and questionnaires. Thus, the interviewee became the conductor of the interview and had the authority to speak about any topic whenever he or she wanted. According to Pereira de Queiroz (1991, p. 71): any information can be utilized to open horizons that the interviewer had not suspected, widening the questioning from the outset, thereby giving it a progressive character. The revelations that the interviewee shows as he thinks about himself, like the formation and inter-relationship of social classes, reveals the domination of groups and social classes and finally how this creates the global society.

The observations and interviews with the children were completed while they were playing games in the morning, the time of day when they were not in school. The questions emerged according to the flow the conversations took. At the outset, we asked the children to teach us their games as they were playing them. In only a few days, groups of children came running in our direction, inviting us to play with them. It was necessary that we participate with all the groups, so we could only spend a little time with each one.

CHILDREN, POVERTY, AND THE STEREOTYPE OF MARGINALITY The poor Brazilian family living in a favela has become part of the modern capitalist world. From this perspective of individualism and modernity, the greatest prejudice is not that of race, gender, or creed, but that directed against poverty or more clearly,

Toys and Games

29

against the poor. However, the large presence of blacks among the disfavored section of the population juxtaposes prejudice against the poor with prejudice against blacks, thereby blending into one attitude. Around one-third of the Brazilian population lives below the lowest consumer level considered minimal by market researchers, in other words, below the levels classified as “C” and “D.” As Kehl (1996) has observed “these other Brazilians don’t appear to belong to the same symbolic order of those who consume, who shop, who spend. In fact, they are not consumers, nor are they citizens.” As a mother living for four years in the favela said, “The poor and those living in favelas have no chance in the world of today.” Society has created a distorted perception of equality between poverty and the marginalized and between poverty and criminality. According to this attitude, to be a resident of a favela intensifies this classification, which Zaluar (1992) has already stated signifies first of all, that to be poor and belong to the excluded classes in our society, “establishes yourself as an object of suspicion to commit illegal or illicit acts, or worse, to be habitually violent.” This is like saying to a father of a family residing in the favela for three years, “People there in the center of town fear the residents of the favela, who they take to be criminal, and would not offer them jobs.” However, the condition of the favela household is not always dominated by unemployment or underemployment. In all the houses visited in this research, the fathers or stepfathers work and, in some cases, the mothers did as well. The majority of mothers worked at home and took care of the children. Nevertheless, the favela worker suffers from low-paid employment and makes up that part of the working class that has as its common characteristic the fact of living in subhuman conditions. Favela residents fill a variety of jobs. Though often unskilled, the favela worker delivers necessary services in industrial plants or in construction. He might only find work on a monthly or even daily basis, or perhaps be self-employed, such as bar owners, renters of small huts, or as collectors of cardboard boxes and plastic for recycling. One father explained, “As much as we work, it won’t help change our life . . . our salary is always finished before the month ends.” This population obtains only the most disagreeable and unstable employment in the marketplace with little or no chance of overcoming the economic plight imposed by a completely insufficient income. Prejudice against favela residents paints a broad stroke. Many fathers, like mothers and even children, referred to the prejudice they suffered in spite of not being unemployed. They would say: “We are not bums.” Jos´e de Souza Martins (1997, p. 16) has argued that to be one of the excluded was “to be in a needy state, but above all to be someone who isn’t recognized as a human being, someone who is stigmatized or considered ominous or dangerous to society.” The children understood the situation that their parents faced, as one ten-yearold boy said in referring to his father’s advice: “My father said that I should not

30

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

let myself be insulted in the street just because I live in a favela without answering back, because I am more honest than most people of this city.” Or in the words of a nine-year-old girl: “My father is poor but he is very honest. He was never a thief.” Street crime, especially violent crime, is one of the main concerns of urban Brazilians today. The image of the city as a violent place generates feelings of fear and insecurity. Today, the city’s image is linked to the deterioration in the quality of urban life. Moreover, the favela residents are represented in society as a group of people who, according to Perlman (1977, p. 28), “exhibit all the symptoms of social disorganization, from the disintegration of the family, anomie, and mutual suspicion to crime, violence, and promiscuity.” It has become common to hear that the favela resident, if not really a criminal, is a potential criminal. They are “inevitably fated to enter the world of crime.” And as this generic label is applied to the residents indiscriminately, all of the favela’s residents are considered dangerous, and if they come in contact with “people who live in areas that have paved streets” are likely to contaminate them with their lack of values. Any contact with favela residents therefore is seen as a threat to the integrity of “good citizens.” Leser de Mello (1999, p. 129) has stated that “all of these prejudices against poor people and favela residents are constantly strengthened by the media: the media is involved in the creation and maintenance of stereotypes and prejudices that stigmatize the poorest people.” On account of these beliefs society stigmatizes individual residents of a favela, those who are considered to live outside society’s norms and are thought to be dangerous. Children living in the favela are often seen as potential vagrants and classified as troublemakers and as little criminals. Some children speak of how they perceive the prejudice against residents of the favela. One eight-year-old boy said: “We know that the people from the city don’t like us here.” A little ten-yearold girl talked about the only time that she had gone to a shopping center with her mother: “When I went there with my mom, they looked at us like we were bandits . . . It seemed that they were afraid of us because we were wearing ugly clothing.” A young boy of eight also referred to the prejudice: “I don’t like living here because other people don’t like those of us who live here.” These comments tell us how the children and other favela inhabitants regard life in the favela and the city; there is the belief that the rest of the city believes that the favela does not represent part of the city and treats it as if it were an inherently antagonistic place. According to Silvia Leser de Mello (1999, p. 135): When we recognize that other people have human qualities we are recognizing that they have the same human attributes that we confirm in ourselves. If we don’t, then we are acknowledging that they are not really fellow human beings. It is difficult for the dominant classes to accept an equal among the class of poor people. They recognize the

Toys and Games

31

difference as inequality, and there is not a great distance to travel from inequality to inferiority.

This hard reality, replete with stigmatizations and prejudices, is the reality for millions of families and of children throughout the country. According to statistics provided by the Brazilian Institute of Geography and Statistics, the number of favelas and the number of shacks within the favelas has increased. This increase can be attributed to another type of exclusion, which many analysts call the “new poor.” The new poor consists of those people who until recently formed part of the labor market, for better or for worse, or they belonged to a family group that was capable of maintaining their integration in the social system even without a job. Now there exists the unemployed worker. In the words of Oliveira de Araujo (1996): This non-class is seen empirically as a combination of workers expelled from work as a result of automation and the information age. Or perhaps this class is a product of the crisis of capitalism and of the dissolution caused by the effect of new production techniques and the new social relations of capitalist production.

Nevertheless, many jobs held by favela residents do provide sufficient income to sustain entire families even though they are not actually recognized formally as jobs. Many times the same family is sustained by various members who hold jobs, or by one family member who holds several jobs. Therefore, when a poor person or a favela resident says that the “poor don’t have a chance,” or that to try something “is not worth it,” this isn’t a sign of resignation or fatalism, but a realistic interpretation of the situation. If the barriers blocking the road were removed, he could respond differently. All the established prejudices are constantly reinforced by the media, and what has been surveyed here provokes a series of questions. What are the social relations established by seven- to ten-year olds with their families? How are infant groups formed? In what way special to the group, do children utilize games and playing? Finally how are gender relations formed when playing games?

TOYS AND GAMES ON THE STREETS OF THE FAVELA From birth, children interact not only with their own bodies and the physical environment, but also with other human beings. As Berger and Berger (1984) have explained the biography of any individual from birth is the history of his relations with other people. Children of the favela Parque das Nac¸o˜ es establish social relations with their families, neighbors, and other favela residents. As we have confirmed through our

32

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

observations and interviews, the children only leave the favela to go to school. However, we have to consider that the favela has specific qualities that other environments don’t have, such as the fact that everyone who lives in the favela is very poor and is socially excluded from mainstream society. The only contact that the children of Parque das Nac¸o˜ es have outside the favela is at school, a place where they can exchange and share different experiences with other children who live elsewhere and with other people, such as teachers, school workers, and the school director. In order to complete research where children are the principal participants, it was necessary for us to determine the setting in which they establish daily social relations. In Parque das Nac¸o˜ es, this world consists of the house and the street. As we have already mentioned, the children in the research were limited to those who attend school in the afternoon. Therefore, we could only conduct our observations and interviews in the morning. During our first visits to the favela, some children exhibited shyness, others curiosity, but in a short time they felt completely at ease. Around nine o’clock every morning (with the exception of rainy days), the children of Parque das Nac¸o˜ es would begin to leave their homes, some carrying toys, others empty-handed, but all with the same desire to play and have fun. In all the streets of the favelas, groups of six or seven neighborhood children play together every morning. The games that children play take on every imaginable form. There are games that utilize toys and those that don’t. Some of these are called “you like this,” “hopscotch,” “hide-and-seek,” or “pass the ring,” “scuffle,” “fat and lazy woman,” “little house,” “little food,” “little school,” and “cops and robbers.” All the names mentioned here are those used by the children themselves. In many games boys and girls play together, much like in hide-and-seek. However, a common thread through these games in the favela is that the children don’t mix. Little girls play “little house” and “little food,” while little boys play “cops and robbers” and soccer. As Brazilian Professors of Psychology Pontes and Magalh˜aes (2003, p. 8) have commented, from the age of three, “children have already demonstrated a preference for playing with members of their own sex, although they do take part in mixed group activities. This segregation is stressed when the child reaches school age.” In Brazilian culture some games are considered appropriate for little girls, such as jumping rape, playing hopscotch, and playing house; while shooting marbles and flying kites are seen as games for little boys. According to Pontes and Magalh˜aes, the most recent data indicated that there is a greater segregation in flying kites than in shooting marbles. Street games enable the observer to witness the gender relations developed by the children. The formation of gender groups by certain

Toys and Games

33

games that then come to be characterized as typically for boys or girls generally leads to a lower status for a member of the opposite sex (p. 8). Helen B. Schwartzman (1995, p. 245) has written about the ethnographic research completed by Marjorie Harness Goodwin into “a group of urban African American children living on ‘Maple Street,’ a working-class neighborhood in southwest Philadelphia”: Girls and boys on Maple Street alternately joined and separated in their play activities, or they might serve as direct or indirect audiences for each other activities. Goodwin documented the variety of ways that boys and girls engaged in talking and playing with each other, such as teasing, storytelling, arguing, playing house, and exchanging ritual insults, and she also discusses how the “ecology of the streets” facilitates a number of cross-sex interactions which are more relaxed and of longer duration than those reported in other studies. This study challenges many assumptions that have influenced child-development researchers, such as the idea of separate gender worlds, which may be apparent in school contexts but not in neighborhood contexts and which also varies by social class and culture (p. 247).

Barrie Thorne (1999, p. 44), an American sociologist, found a situation similar to the Parque das Nac¸o˜ es favela streets, when researching four- and five-year-old children at two North American primary schools. She observed that whether during classes, while walking in the school corridors, or while eating lunch, the children generally did not form groups by gender. However, the games played at the school playground were strongly separated by sex. Thus, the little boys played soccer and baseball together, while the little girls skipped rope and did simple gymnastics in a paved area of the playground. Boys and girls generally got together only to play “hide and seek” or handball. All of this research above leads to the conclusion that is not one hard fast rule in the formation of playgroups by sex. Even within the same society, similarities and differences could exist among the groups of children studied, depending on their location, their social class and their local culture. Comparing the games played by Brazilian children in the Parque das Nac¸o˜ es favela with those played by North American children researched by Barrie Thorne there are some obvious similarities among the games that join boys and girls together and those that separate them by sex. Mothers can often serve as helpers in the games that children play as well as acting as disruptive intruders. The mother and child relationship is often connected to the family’s socio-economic status, the parents’ education level, and to the age and sex of the children (Youngblade & Dunn, 1995, pp. 227, 228). Brazilian society is deeply marked by its patriarchal family relations, although it has modernized economically. These patriarchal relationships are still widely found among the lower social classes (Kuznesof, 1988/1989). Thus we should

34

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

not be surprised by the observations that we make in regard to the mother-child relationship and to the classification of toys as either masculine or feminine. Many times, mothers strongly influenced the distinctive games that boys and girls played. In conversations between mothers and sons, we noted that mothers often would not permit their sons to play with dolls with their sisters or neighbors. In contrast, mothers approved of their daughters playing soccer or playing with small carts. The fact that dolls are culturally identified as girls’ toys made many mothers consider playing with dolls to be a sign of homosexuality and led many boys to feel the same way. The following is a brief dialogue that took place on a favela street: Mom: I’ve already told you not to play with these things. Son: Mom, but Tati called me to play. Mom: Tell Tati to call her girlfriends to play. This is a girl’s thing. Son: Oh, mom! Let me play.

Soon after this conversation, we observed this young boy crying on a street in the favela and his mother telling him to play inside the house with his little trucks. This wasn’t the only time that we witnessed mothers preventing their boys from playing with dolls because dolls were only for girls. The grandmothers also thought like this and didn’t like their grandsons playing with dolls. On the other hand, girls could play with masculine games without their mothers or grandmothers objecting. Mothers and grandmothers believed that boys playing with dolls would lead neighbors and other people in the favela “to make jokes in bad taste about the boys being homosexual.” In her research, Barrie Thorne (1999, p. 131) observed that the children themselves differentiated games and toys by sex. Boys who tried to play with girl’s games were rejected by the girls and called sissy by the other boys. However, the little girls who crossed the gender barriers and showed athletic skills or verbal ability were accepted by the little boys. In their own world, little girls could play with dolls and create different situations for their games, such as a doll being a woman who worked in an office, or as a doctor. We often questioned the little girls about the occupation of the doll that they were playing with and they responded that the profession of the doll was one that they themselves would like to have when they became adults, or “when I am big,” using the children’s own words. Researcher: Why is your doll a secretary? A little girl: Because she works with a computer and computers are good. I am going to be a computer operator when I grow up and I am going to be a secretary (When we were asking this question, this little girl was playing office with another girl). Researcher: Your doll is a doctor, are you playing hospital?

Toys and Games

35

A little girl: No, I’m playing how to cure sick people. Carol and I are going to become doctors when we grow up, but we could also be nurses wearing white uniforms (Carol is a friend who responded to the question).

Some little girls also expressed a desire to be like Xuxa or Eliana, who host children’s television shows, and talked of having the toys that they had seen on Eliana’s television show. “I want to be like Xuxa . . . she’s very pretty.” “I’m going to be just like Eliana on Channel Record, you know her?”

Gilles Brougere (2000, p. 37), the French educator, in his book, Games and Culture, said that “The world of the Barbie doll or of Xuxa are two ways that little girls express their aspirations and their desires.” Through observing games we perceived the great sociability that already exists among the children in the Parque das Nac¸o˜ es favela. During the morning, children talk a lot and tell each other stories about what happened at school or about other things in their homes. From these conversations between children we learned many details about their daily life at home and at school and their ideas about life, the world, and the future. The following dialogue demonstrates the socialization that exists within the home, which spills over into children’s play. Girl # 1: My father is going back to study. Girl # 2: That’s great! Girl # 1: Let’s play school. Now, I’m going to play my father and you can be the teacher.

Soon afterward, the little girls started playing school, designing a blackboard on the dirt street and handing out some sheets of paper and pencils. We watched the following scene that took place during the class. Girl # 1: Do this lesson. Mr. Zeca. Girl # 2: Yes. That’s good. I’m doing it. Girl # 1: Very good. I liked it a lot. You’ve passed.

Thus, we can see how many aspects of social relations that take place at home are used as themes of infant play. Games are also passed down from mothers to children. In the Parque das Nac¸o˜ es favela, a majority of mothers don’t work and as a result the children develop a more intense sociability with their mothers than with other family members. Thus, mothers teach their children games and they build toys based on their own memories as children. Children use their toys imaginatively, playing out their own desired reality. Our research shows that children living in the favela, much like the children of Bom Retiro, a lower-middle-class S˜ao Paulo neighborhood studied by the noted

36

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

Brazilian sociologist Florestan Fernandes (1979) in the 1940s, created their own games and toys. In Bom Retiro, Fernandes concluded that by playing games with each other, children developed their artistic sense as well as their peer group relationships. Toys created by parents for children also have special value. In the favela we found small wooden toy carts, paper boats, dolls, and doll clothing that had been designed by mothers. We also found mothers and grandmothers teaching games from their generations to the children. According to Brougere (2000, p. 12): Artisan toys will always have a very important role in the social relations of people. They are irreplaceable precisely because they are completely designed by men, not by machines, as part of the human rhythm of life, as a product of manual skill, of fantasy and of the capacity of each person to create something.

Relations formed in school also frequently reflect children’s games. It is easy to find children playing “little school” on the streets of the favela. It is in the real school that the children can contrast life with how it actually is on favela streets. At school they get together with people from other neighborhoods and this fact is regularly reflected in their games. The references that children make to school always emphasize that the building, the teacher, and the other people comprise a world far from their own reality, as if everything in the school, both the people and the physical space, were more beautiful than anything in the favela. The school building was recently built and perhaps that is why the children made so many references to the building. In addition, the children’s attention to beautiful people is a result of the attractive clothing that the teachers and school staff use, in contrast to the clothing worn by their own mothers and fathers in the favela. The children of the favela see their school and its workers as attractive, and their answers to questions about the value they attach to their studies indicate that nearly all the children want to attend college with the exception of some little boys who dream of becoming soccer players. In their conversations, these boys show that they believe in the idea of becoming great players and the media reinforces this daily by presenting millionaire players who were formerly residents of a favela or at least had a poor childhood. Some boys of the Parque das Nac¸o˜ es favela envisioned themselves in these examples of social mobility and dreamed of one day also having such an opportunity. As for the girls, fully 100% of them had a desire to attend college. These desires for the future are typical of childhood games. In their improvised games they play at being doctors, dentists, judges, teachers, engineers, and they also play at workplaces that have computers. Even in situations that threaten their own survival, the children in the favela don’t appear to be marked by hardship. They show themselves to be capable of

Toys and Games

37

creating elaborate games and of dreaming of a more worthy future, even though it will be very difficult to achieve these dreams. There are also games of cops and robbers in which we see children using their own fingers to make the outline of a gun. However, we want to emphasize that the games of cops and robbers are not violent. They appear to be rather innocent, only reproducing the scenes of shows that they have watched on television. Brougere (2000, p. 88) has emphasized that: . . . it is necessary to admit that using a weapon as a toy is not generally a violent act, but only a representation, perhaps necessary, of a violent act which . . . appears to be healthy because through his games, the child discovers the many aspects of his culture.

In this way, the games in which children play with weapons are ways by which they experiment with the reality that they watch on television or hear about in stories. The question of heroes is another point to be raised. The children already regard the police and the good guys highly because they want to be just like them, dispensing justice and acting like good people, a concept that is emphasized by their mothers in their education.

DISCUSSION We have observed that the children living in the Parque das Nac¸o˜ es favela have at their disposal fewer store-bought toys than do middle-class children. Nevertheless, the favela children have more opportunity to play with neighborhood children, on local streets than middle-class children. The separation of toys and games by gender owes much to the influence of mothers and grandmothers. The mothers’ contact with their children is intensified because they generally do not work and stay at home. This gender distinction of games and toys could also be connected to the family’s socio-economic status and to the parents’ level of education. The children’s games contain elements of their daily life, including their access to television and school. It is at school, where these children face middle-class people and have to cope directly with the questions of marginalization and social exclusion. Yet, these contacts do not prevent them from exercising their creativity and imagination when they are playing with the possibility of a better life.

CONCLUSION We have concluded that through their games, the children have demonstrated their dreams, fears, anxieties, and especially their views of daily life that are so important

38

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

for our research about the sociability of children who suffer from social exclusion. The ways in which poverty, indifference, and stigmatization afflicts children also afflicts society at large, but that is a subject for other research. Sociability among infants is of primary importance in the life of a child, whether it is sociability with other children, with the family, or at school. Through this sociability, the child gains knowledge of the world around him. Children of the subordinate classes, according to Berger and Luckmann (1974), not only absorb the perspective of their own class with respect to the social world, but they also absorb this perception with a particular coloring that their parents, or any individuals in charge of their primary socialization, provide them. Thus, there are many opinions that favela children themselves possess and the situations in which they find themselves are built through the social relations established in the family. As a result of conversations with the families within the favela, we saw how opinions based on common opinion with respect to the residents of the favela are distorted. Indeed, the families residing in the Parque das Nac¸o˜ es favela take great care of their children and are concerned about their future. This contrasts with the stereotypical notion that poor families and those in favelas are negligent in caring for their children’s education. As we have seen, children utilize their games played on the street, their own creativity, and their learning of old games from their mothers and grandmothers in order to find alternatives to experience this important phase of life: childhood.

NOTES 1. Some of the works consulted on the topic of Psychology of Development include: Pontes and Magalh˜aes (2003), Sager and Sperb (1998). Some of the works consulted on the topic of Education include: Kishimoto (2001), Zancope (2003). 2. Some of the works written in the field of Sociology include: Machado Neto (1979, 1980), Kosminsky (1991), Ferreira (1980), NEPI (N´ucleo de Estudos e Pesquisas sobre a Infˆancia) (1989), Volpi (1997), Nogueira, Romanelli and Zago (2000), Landini (2000). 3. Some of the works written in the field of Anthropology include: Ferraz (1991), Oliveira (1999), Gregori (2000), Fonseca (1985). 4. Some of the research supported by International Labour Office (ILO), Organizac¸a˜ o Internacional do Trabalho (OIT), in Portuguese: Saboia (2000), Rizzini and Fonseca (2002). Some of the research about child labour at Brazilian universities: Carneiro (2002), Kosminsky, Angelin, Santana and Rodrigues (2002). 5. Violante (1985). 6. Abramo (1994). 7. Vianna (1988), Abramovay et al. (1999). 8. Marc´ılio (1998), Venˆancio (1999). 9. Cascudo (1984) [1978].

Toys and Games

39

10. One of the most important Brazilian authors is Monteiro Lobato who started writing for children in the 1920s. Through his books it’s possible to recover toys and plays from a time when most of the Brazilian population lived in rural areas. Some of his books became children’s TV plays. Helena Morley wrote a very interesting diary between 1893 and 1895 when she was a young girl living in a small town in the Brazilian countryside.

REFERENCES Abramo, H. W. (1994). Cenas juvenis: punks e darks no espet´aculo urbano. S˜ao Paulo: Scritta, ANPOCS. (See the Introduction). Abramovay, M., et al. (1999). Gangues, galeras, chegados e rappers: Juventude, violˆencia e cidadania nas cidades da periferia de Bras´ılia. Rio de Janeiro: Garamond. Bastide, R. (1979 [1947]). Pref´acio, as “Trocinhas” do Bom Retiro. In: F. Fernandes (Ed.), Folclore e Mudan¸ca Social na Cidade de S˜ao Paulo (pp. 153–155). Petr´opolis: Vozes. Berger, P. L., & Berger, B. (1984). Socializac¸a˜ o: como ser um membro da sociedade. In: M. M. Foracchi & J. S. Martins (Eds), Sociologia e Sociedade (pp. 200–214). Translated by R. P. Neto. Rio de Janeiro: Livros T´ecnicos e Cient´ıficos Editora S. A. Berger, P. L., & Luckmann, T. (1974). A constru¸ca˜ o social da realidade. Translated by F. de S. Fernandes (2nd ed.). Petrop´olis: Editora Vozes Ltda. Brougere, G. (2000). Brinquedo e Cultura. Translated by M. A. de S. Doria (3rd ed.). S˜ao Paulo: Cortez Editora. Carneiro, A. G. D. (2002). Erradicac¸a˜ o do trabalho infantil: Estudo de pol´ıticas p´ublicas de combate a` explorac¸a˜ o da m˜ao-de-obra infantil em Curitiba. Dissertac¸a˜ o de Mestrado, Programa de P´osGraduac¸a˜ o em Sociologia, Universidade Federal do Paran´a. Not published. Cascudo, L. C. (1984) [1978]. Literatura oral no Brasil (3rd ed.). Belo Horizonte: Itatiaia; S˜ao Paulo: EDUSP (See Chapter II and IV). Fernandes, F. (1979 [1947]). As “Trocinhas” do Bom Retiro. In: F. Fernandes (Ed.), Folclore e Mudan¸ca Social na Cidade de S˜ao Paulo (pp. 153–258). Petr´opolis: Vozes. Ferraz, I. (1991). Os ´ındios Parkatejˆe 30 anos depois. In: J. de Souza Martins (Ed.), O massacre dos inocentes: A crian¸ca sem infˆancia no Brasil. S˜ao Paulo: Hucitec. Ferreira, R. M. F. (1980). Meninos de rua – valores e expectativas de menores marginalizados em S˜ao Paulo. S˜ao Paulo: Cedec/Comiss˜ao de Justic¸a e Paz. Fonseca, C. (1985). Valeur marchande, amour maternel et survie: Aspects de la circulation des enfants dans um bidonville Br´esilien. Annales ESC, (5), 991–1022. Gregori, M. F. (2000). Vira¸ca˜ o: Experiˆencias de meninos nas ruas. S˜ao Paulo: Companhia das Letras. Kehl, M. R. (1996/1997). Vocˆe decide e Freud explica. In: J. Lerner (Ed.), O Preconceito (pp. 63–75). S˜ao Paulo: Imprensa Oficial do Estado. Kishimoto, T. M. (2001). Brinquedos e materiais pedag´ogicos nas escolas infantis. Educa¸ca˜ o e Pesquisa, 27(2), 229–245. Kosminsky, E. (1991). Internados – os filhos do estado padrasto. In: J. de Souza Martins (Ed.), O massacre dos inocentes: A crian¸ca sem infˆancia no Brasil. S˜ao Paulo: Hucitec. Kosminsky, E., Angelin, P., Santana, J. N., & Rodrigues, J. A. (2002). O trabalho infanto-juvenil dom´estico na cidade de Mar´ılia. An ongoing research project started in 2002. supported by PIBIC/CNPq/UNESP.

40

ETHEL VOLFZON KOSMINSKY AND LAURA DANIEL

Kosminsky, E. V. (1992). Procedimentos metodol´ogicos e t´ecnicos na pesquisa com crianc¸as assistidas. In: A. B. S. G. Lang (Ed.), Reflex˜oes Sobre a Pesquisa Sociol´ogica (pp. 61–77). S˜ao Paulo: Textos CERU (3). Kuznesof, E. A. (Sep. 1988/Feb. 1989). A fam´ılia na sociedade Brasileira: Parentesco, clientelismo e estrutura social (S˜ao Paulo, 1700–1980). In: E. de Mesquita Sˆamara (Ed.), Fam´ılia e Grupos de Conv´ıvio, Revista Brasileira de Hist´oria, 9(17), 37–63. Landini, T. S. (2000). Pornografia infantil na internet: Prolifera¸ca˜ o e visibilidade. Dissertac¸a˜ o de Mestrado, Programa de P´os-Graduac¸a˜ o em Sociologia. S˜ao Paulo: USP. 157 pp. Not published. Leser de Mello, S. A. (1999). A violencia urbana e a exclus˜ao dos jovens. In: B. Sawaia (Ed.), As ´ Artimanhas da Exclus˜ao: An´alise Psicossial e Etica da Desigualdade (2nd ed., pp. 129–140). Petr´opolis: Editora Vozes. Machado Neto, Z. (1979). Meninos trabalhadores. Cadernos de Pesquisa, (31), 95–102. Machado Neto, Z. (1980). As meninas: sobre o trabalho da crianc¸a e do adolescente na fam´ılia prolet´aria. Ciˆencia e Cultura, 32(6), 671–683. Marc´ılio, M. L. (1998). Hist´oria social da crian¸ca abandonada. S˜ao Paulo: Hucitec. Martins, J. S. (1997). Exclus˜ao e a nova desigualdade. S˜ao Paulo: Paulus. Mergen, B. (1995). Past play: Relics, memory, and history. In: A. D. Pellegrini (Ed.), The Future of Play Theory: A Multidisciplinary Inquiry into the Contributions of Brian Sutton-Smith (pp. 257–274). Albany: State University of New York Press. Morley, H. (1998 [1942]). Minha vida de menina. S˜ao Paulo: Companhia das Letias. NEPI (N´ucleo de Estudos e Pesquisas sobre a Infˆancia) (1989). A infˆancia violada: um recorte atual. Rio de Janeiro: LPS/UFRJ. Nogueira, M. A., Romanelli, G. & Zago, N. (Eds) (2000). Fam´ılia e escola: Trajet´orias de escolariza¸ca˜ o em camadas m´edias e populares. Petr´opolis: Vozes. Oliveira, I. (1999). Desigualdades raciais: Constru¸co˜ es da infˆancia e da juventude. Niter´oi: Intertexto. Oliveira de Araujo, M. N. (1996). Novos pobres: o que h´a de novo? Revista de Ciˆencias Sociais, 27(1/2), 85–98. Pereira de Queiroz, M. I. (1991). Varia¸co˜ es sobre a t´ecnica de gravador no registro da informa¸ca˜ o viva. S˜ao Paulo: T.A. Queiroz, Editor. Pereira de Queiroz, M. I. (1992). Desenvolvimento das ciˆencias sociais no Brasil: Nascimento e expans˜ao. Revista Ciˆencia e Tr´opico, 20(2), 387–412. Perlman, J. E. (1977). O mito da marginalidade: Favelas e pol´ıticas p´ublicas no Rio de Janeiro. Translated by W. M. Portinho. Rio de Janeiro: Editora Paz e Terra. Pontes, F. A. R., & Magalh˜aes, C. M. C. (2003). A transmiss˜ao da cultura da brincadeira: algumas possibilidades de investigac¸a˜ o. Psicologia: Reflex˜ao e Cr´ıtica, 16(1), 117–124. Portes, A. (2002). Sociology in the hemisphere: Convergences and a new conceptual agenda. Latin America Studies Association, XXXIII(1), 6–7. Rizzini, I., & Fonseca, C. (2002). As meninas e o universo do trabalho dom´estico no Brasil: Aspectos hist´oricos, culturais e tendˆencias atuais. Rio de Janeiro: OIT/IPEC. Saboia, A. L. (2000). As meninas empregadas dom´esticas: uma caracteriza¸ca˜ o s´ocio-econˆomica. Rio de Janeiro: OIT/IPEA. Sager, F., & Sperb, T. M. (1998). O brincar e os brinquedos nos conflitos entre crianc¸as. Psicologia: Reflex˜ao e Cr´ıtica, 11(2), 309–326. Schwartzman, H. B. (1995). Representing children’s play: Anthropologists at work. In: A. D. Pellegrini (Ed.), The Future of Play Theory: A Multidisciplinary Inquiry into the Contributions of Brian Sutton-Smith (pp. 243–256). Albany: State University of New York Press. Thorne, B. (1999). Gender play: Girls and boys in school. New Brunswick: Rutgers University Press.

Toys and Games

41

Venˆancio, R. P. (1999). Fam´ılias abandonadas: Assistˆencia a` crian¸ca de camadas populares no Rio de Janeiro e em Salvador – s´eculos XVIII e XIX. Campinas: Papirus. Vianna, H. (1988). O mundo funk carioca. Rio de Janeiro: Jorge Zahar. Violante, M. L. (1985). O dilema do decente malandro. S˜ao Paulo: Cortez, Autores Associados. Volpi, M. (1997). O adolescente e o ato infracional. S˜ao Paulo: Cortez. Youngblade, L. M., & Dunn, J. (1995). Social pretend with mother and sibling: Individual differences and social understanding. In: A. D. Pellegrini (Ed.), The Future of Play Theory: A Multidisciplinary Inquiry into the Contributions of Brian Sutton-Smith (pp. 221–240). Albany: State University of New York Press. Zaluar, A. (1992). Exclus˜ao social e violˆencia. In: Sociedade Civil e Educa¸ca˜ o (pp. 113–123). Coletanea CBE (Cedes), Campinas: Editora Papirus. Zancope, L. C. (2003). O tempo da crian¸ca na pr´e-escola e o lugar do brincar. Monografia de conclus˜ao do Curso de Padagogia. Mar´ılia: UNESP. 38 p. Not published in press.

TRANSITION TO ADULTHOOD IN JAPAN AND KOREA: AN OVERVIEW Hyunjoon Park and Gary D. Sandefur INTRODUCTION A good deal of work in demography and sociology has examined the dynamic and complex patterns of the process through which young people move from being dependent on their parents and families to supporting themselves financially and forming their own families. These studies have described, in detail, specific features of many aspects of the transition to adulthood, especially in the European or American context (Corijn & Klijzing, 2001; Hogan & Aston, 1986; Sandefur, Eggerling-Boeck & Park, forthcoming). They have, furthermore, provided us with some explanations of how institutional variations in educational systems, labor markets, or family formation are associated with differences in European and American young people’s transitions to adulthood (Breen & Buchmann, 2002; Cook & Furstenberg, 2002; Fussell, 2002a). However, in order to have a better understanding of how cultural or institutional variations across countries explain national differences in the patterns and timing of the transition from youth to adulthood, we need to extend our interest into other regions that have very different institutional arrangements and cultural contexts from those of European or American countries. By examining the extent to which the experiences of young people in those areas are similar to or differ from the experiences of becoming an adult for European or American youth, we can obtain a better understanding of how various determinants of the transition to adulthood may operate differently, depending on the specific context. Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 43–73 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10004-4

43

44

HYUNJOON PARK AND GARY D. SANDEFUR

Among other regions, two East Asian countries, Japan and South Korea, are particularly interesting in terms of their cultural traditions and institutional arrangements and thus provide an excellent case to which to compare European and American experiences. During the last few decades, the two East Asian countries have experienced remarkably rapid social, economic, and demographic changes. We are interested in exploring how institutional characteristics have shaped the patterns of the transition to adulthood among Japanese and Korean youths and how they differ from those in other societies. Furthermore, we address how the patterns have changed during the last few decades in the context of rapid changes across various dimensions of social structure.

THEORETICAL APPROACH The transition to adulthood can be viewed from a number of different and complementary perspectives, including sociology, developmental psychology, biology, and demography. In this paper we take a predominantly social demographic approach and look at the transition to adulthood in terms of four key markers: the completion of education, entering the labor force, family formation, and leaving home. These four markers have been studied a great deal in the American and European contexts. One way to view the early transition to adulthood is as a sequence of decisions that individuals make to prepare themselves for full adult status. Research suggests that social structure and cultural traditions create a set of opportunities and constraints within which people make decisions. Japanese, Korean, and U.S. social structure and cultural traditions vary in many ways. Within their social structural and cultural contexts, we assume that individuals attempt to enhance their shortterm and long-term well-being, and that they base their decisions on the perceived costs and benefits of alternative choices. In other words, agency as well as cultural traditions and social structure, plays a key role in the transition to adulthood. For example, when it comes to deciding whether to work or enter post-secondary education after high school, what we know about the financial benefits of attending post-secondary education suggests that this is the best choice. But for some people the costs of post-secondary education may be too high for them or their families to bear, or the combination of an immediate income and the lower costs associated with working may make it a more attractive option than going to school. The costs of pursuing post-secondary education also vary with the type of institution and whether or not the student lives away from home while attending school. Of course people sometimes make poor choices. This may be the result of bad information, for example, in the case of someone who thinks the long-run

Transition to Adulthood in Japan and Korea

45

benefits of going to work immediately are higher than those of completing postsecondary education and then working. Or, someone may be unfamiliar with financial aid and other ways of reducing the cost of post-secondary education. Having a child out-of-wedlock early in life is often due to a decision to have sex without contraception, resulting in an unintended pregnancy that is carried to term. Within this framework, completing education, entering the world of work, family formation, and leaving home emerge as key decisions that affect the wellbeing of individuals not only in the immediate post-adolescent period but well into their later adult years. We are interested in both the prevalence of these markers at different ages in late adolescence and early adulthood, and the possible institutional and cultural forces that might account for some of the differences between the transition to adulthood in the United States and East Asia. We are not in a position to try to establish a causal connection between institutions and culture on the one hand and these demographic markers on the other, but we propose some possible connections based on the national data that are available. We begin with a brief review of some of the cultural and institutional forces in Japan and Korea, which establishes a context within which the transition to adulthood occurs. We then review secondary data sources that provide information on the four key markers: education, labor force activity, family formation and leaving home. Finally, we conclude with a discussion of how research might investigate more thoroughly the sources of differences in the American and East Asian experiences.

CULTURAL AND INSTITUTIONAL FORCES SHAPE THE TRANSITION TO ADULTHOOD The common Confucian cultural traditions of Japan and Korea establish some norms about education, work, family formation, and responsibility to parents. The latter are related to if and when children expect to leave their parental home. In addition, these societies have educational and social welfare systems that differ from those in the United States in significant ways that likely affect the transition to adulthood.

Cultural Norms The common underlying Confucian culture in East Asian societies plays an important role in many aspects of the transition to adulthood. Contemporary East

46

HYUNJOON PARK AND GARY D. SANDEFUR

Asian young people do not frequently discuss Confucian ideas, yet this tradition continues to exert an influence over norms and behavior that is perhaps even stronger than the influence of the Judeo-Christian cultural norms and values in the United States. In particular, a strong patriarchal culture emphasizing women’s childbearing roles and their family responsibilities tends to devalue women’s economic activities. Korean women show one of the lowest levels of labor force participation among the Organization for Economic Co-operation and Development (OECD) nations and it is well known that in Japan and Korea, a large portion of women stop working and exit the labor market upon marriage or childbearing (Brinton, 2001). The Confucian tradition in East Asia is an important factor affecting not only young women’s economic activities and family expectations but also influencing young people’s overall behaviors and attitudes. For example, the culture has a strong social norm, often gender-biased, in regard to risk behaviors such as smoking or drinking. It requires young people to treat seniors or elderly in specific ways. The culture also has a strong norm of labeling premature sex or childbearing as wrong, and this influences young people’s sexual activity outside marriage.

The Educational System Both Japan and Korea experienced a great deal of economic development during the last half of the 20th Century, although the period of remarkable economic growth in Japan was earlier than in Korea. As industrialization evolved, the nature of employment in these countries also changed dramatically. An even more relevant social change in the two countries regarding the transition to adulthood is their dramatic educational expansion, as we will see later. As many studies have pointed out, the expansion of the educational system is a major structural change with respect to the transition to adulthood, affecting other important transition events. As young people remain in school longer, they tend to postpone family formation given that enrollment in the educational system is more or less incompatible with forming their own family (Blossfeld & Huinink, 1991; Corijn & Klijzing, 2001). Depending on the organization of post-secondary education, for instance, whether grants or subsidies are available for a student to leave their home and pursue tertiary education in other areas, the expansion of higher education also affects the age at leaving home (Breen & Buchmann, 2002). In addition to enrollment in the educational system, the institutional characteristics of the educational system such as the degree of standardization and stratification of the educational system have been found to affect the young people’s

Transition to Adulthood in Japan and Korea

47

transition from school to work, which is an important aspect of the transition to adulthood (Allmendinger, 1989; Shavit & M¨uller, 1998). Japan and Korea share the same basic educational structure: 6 years of elementary school, 3 years of middle (lower secondary) school, 3 years of high (upper secondary) school, and 4 years of university (or 2–3 years of junior college). Tracking is not done in lower secondary education while at the upper secondary level schools are differentiated between general (academic) or vocational tracks. The two countries also have a very high level of standardization in that under the direct control of the government, teachers’ training, school budgets, and even the number of college students are constrained by government guidelines (Ishida, 1998; Park, 2004). A nationwide entrance examination for high schools and colleges, and a common curriculum designed to prepare students for the entrance examination reflect the high level of standardization of educational systems in these societies, which is contrasted to more localized and less standardized systems in the United States. In cross-national studies of young people’s transition from school to work, the literature has grouped countries on the basis of their dominant upper secondary pathways, given the relevance of a pathway for labor market or tertiary education (OECD, 2000). According to this classification, Japan and Korea are typical countries with a “general education” pathway where more than 50% of high school students enroll in general (or academic) rather than vocational programs. Specifically, about 70 and 60% of Japanese and Korea upper secondary students, respectively, enroll in general (academic) high schools designed mainly to prepare students for tertiary education and the remaining 30% and 40% of students go to vocational high schools that prepare students for the labor market. The two East Asian countries are clearly in contrast to apprenticeship countries like Germany or Switzerland where more than half of upper secondary students participate in apprenticeship-type arrangements. The excess demand for university education among high school students in Japan and Korea is well known. In both countries, student selection for tertiary education is mainly based on test scores on the national entrance examination. Students are sorted into higher education hierarchically ordered, depending on their test scores. Education in high schools is basically to prepare students for the test, and the high school’s success rate of sending students to top universities has been considered the most important indicator of educational quality the school produces. In this situation, the high school curriculum concentrates on preparation for entrance examinations and students suffer from intense pressure and competition in relation to the examinations. Substantial proportions of students spend a couple of years on preparation for the entrance examination even after finishing high school in order to achieve better scores by retaking the exam.

48

HYUNJOON PARK AND GARY D. SANDEFUR

The Social Welfare Systems Another important institutional factor relevant for cross-national differences in the transition to adulthood is the state’s involvement in providing social welfare programs. For instance, on the basis of Esping-Andersen (1990)’s work on welfare regimes, Breen and Buchmann (2002) outlined four types of welfare arrangements (conservative, southern European variant of conservative cluster, liberal, and social democratic welfare regimes) among Western societies and linked variations in welfare regimes to national or regional differences in several aspects of transition to adulthood including youth unemployment, family formation, and behavior problems. Japan and Korea in general have a conservative social welfare system with a very low level of spending by the government on what we might broadly think of as social welfare, including public education.1 For example, in 1994 the percentage of GDP spent by governments on educational institutions was 3.8 and 3.6% in Japan and Korea, respectively (OECD, 1997). This level is significantly lower relative to over 6.5% in the Scandinavian countries. The most evident indicator of the relatively lower levels of governmental subsidies to education in Japan and Korea is the relative proportion of funds for tertiary educational institutions from public and private sources. In Korea, the relative share of private funds for tertiary education was 83% in 1994, which was the highest among the 16 OECD countries providing private expenditure data. Indeed, except for Korea, only two countries, Japan (54%) and the U.S. (52%) had more than half of the expenditure on tertiary education funded by the private sector, while in other countries most of the expenditure on tertiary education is from public sources. Coupled with the strong cultural norm emphasizing young people’s responsibility for taking care of their parents, this low level of welfare provision by the state probably had a significant influence on young people’s decisions about leaving home and living arrangements in East Asia. Because they financially depend on their family’s support for their education, it is quite common for young people to stay with their parents until they finish tertiary education. After entering labor markets and even marriage, some young people, especially first sons and their wives, tend to live together with their parents, taking care of them. Therefore, the generally less generous social provision in East Asian than in other Western industrial societies is coupled with a strong reliance on the family as a basic unit of economic provision. As we have seen so far, Japan and Korea have very distinctive features from European or American societies across various institutional dimensions like the educational system, labor market, welfare provision and cultural norms. This provides a good opportunity for a comparative study to understand how

Transition to Adulthood in Japan and Korea

49

institutional variations are associated with different patterns of the transition to adulthood. Of course, our focus on the factors similarly underlying Japan and Korea should not lead us to neglect substantial differences between the two societies. As the data show in the following sections, each country has its own distinct patterns.

USING PUBLISHED DATA TO EXPLORE THE TRANSITION TO ADULTHOOD Taking into account specific institutional arrangements, in this overview we describe several aspects of the transition to adulthood among Japanese and Korean young people compared to American young people. For this purpose, we rely on published data. We also introduce some information from representative countries of Western Europe to give a broader comparative perspective on East Asian young people.2 The transition from youth to adulthood is not a single process. Instead, many sociologists and social demographers have approached the study of this period by looking at various “markers” of the transition to adulthood (Hogan, 1978; Hogan & Aston, 1986; Marini, 1984; Modell, Furstenberg & Hershberg, 1976; Rindfuss, Swicegood & Rosenfeld, 1987; Sandefur, Eggerling-Boeck & Park, forthcoming). These markers include various events on the life course such as completing full-time education, entering the labor market, living away from home, marriage, and having children. Drawn from those previous literatures, we particularly focus our attention on four markers of the transition to adulthood: (1) educational achievement and attainment; (2) labor market outcomes including employment and unemployment; (3) family formation including first marriage and childbearing; and (4) leaving home. For each of these four aspects of the transition from youth to adulthood, we describe the country-specific patterns and the timings at which each transition event occurs to identify the similarities or differences between on the one hand the two East Asian countries and other American and European countries on the other. As far as the data are available, we document how the patterns and timings of the transition to adulthood have changed in each society. Furthermore, special attention is paid to gender differences in each aspect of the transition to adulthood within a country and how the patterns of gender difference vary by countries.

EDUCATION As a great deal of social science research has demonstrated, educational attainment is one key to much of what happens to people later in their lives (Hogan & Aston,

50

HYUNJOON PARK AND GARY D. SANDEFUR

1986). The amount and quality of education that a person achieves affects her occupation and career, her lifetime earnings, income, and wealth accumulation, and whom she marries as well as many other features of early, mid-, and late life. In this regard, the overall level of education in a society is a key indicator of the social and economic well being of the country.

Educational Achievement International comparisons of student achievement have been considered useful for assessing how each country’s educational system functions. In various international tests of student achievement, Japanese and Korean students have shown significantly higher levels of achievement than those in other regions. For instance, in the Third International Mathematics and Science Study (TIMSS), Korean fourth and eighth graders scored significantly higher than students from any other countries. Except in only the science performance of eighth graders where they were ranked third, Korean students showed the highest average scores for both subjects. Japanese fourth and eighth graders showed the second highest achievement for both math and science (OECD, 1997). Figure 1 presents the average rank in five different measures indicating the percentage of 14- to 15-year-olds who fall below fixed international benchmarks of competence in reading, mathematics, and science from two separate surveys of the PISA (Programme for International Student Assessment, 2000) and TIMSS (1999).3 That is, by combining five separate indicators of educational disadvantage from the two recent international studies of educational performance, the figure provides a stable and broad overview of the extent to which each country’s educational system is effective in preparing students for the future (UNICEF, 2003). The figure basically indicates that Korea (average rank of 1.4) and Japan (average rank of 2.2) have much lower proportions of students who cannot reach fixed benchmarks of educational achievement and hence they do a better job than others in educational performance. In contrast, the southern European countries like Portugal, Greece, Italy, or Spain occupy the bottom rank with Germany and the U.S. positioned at the next bottom level. Japan and Korea are distinctive in not only their higher mean scores of achievement tests but also their smaller differences in achievement between students. Japanese 15-year-old students displayed the highest mean scores in mathematic literacy of PISA, while the corresponding Korean students showed the highest average performance in science literacy. Further, students in these two countries demonstrated the narrowest gaps in achievement scores between students, as indicated by the smallest differences between the 75th and 25th

Transition to Adulthood in Japan and Korea

51

Fig. 1. Average Rank in Five Measures of the Percentage of Children Scoring Below a Fixed International Benchmark of Educational Performance. Note: The five measures used to create the rank are those from the following international surveys: reading, mathematical, and scientific literacy of 15-years-olds in PISA 2000 and mathematics and science achievement of eighth graders in TIMSS 1999. Source: UNICEF (2003).

percentiles on the mathematical and scientific literacy scales among the countries in PISA (OECD, 2001a). Indeed, Korea and Japan are among a few countries that attain a high level of overall performance and the narrowest disparities in student performance at the same time. On the other hand, Germany and Greece are among a few countries showing the widest variation with low mean scores. In regard to gender differences in student achievement, however, Japan and Korea show a different pattern. In contrast to the overall high mean score with the smallest gap in student performance, Korea shows the largest gender difference in mathematical and scientific literacy among the countries in PISA, favoring males.4 On the other hand, Japanese male students exceed female students much less in mathematical literacy than do Korean male students. Even Japanese females do better than males in scientific literacy, though the difference is not large enough to be statistically significant (OECD, 2001a).5 Educational Attainment In many countries, upper secondary education functions as preparation for further studies in tertiary education or for directly entering labor markets. In this respect,

52

HYUNJOON PARK AND GARY D. SANDEFUR

Fig. 2. Upper Secondary Graduation Rates by Gender. Source: OECD (2001b).

the upper secondary graduation rate indicates the extent to which a country’s educational system is succeeding in equipping individuals with the minimal qualifications needed to make a successful transition to the world of work, a key component of adult life. Figure 2 presents upper secondary graduation rates by gender in 1999 for selected countries with the average rate for the 20 OECD countries reporting the data in the two far-right hand columns (OECD, 2001b). The rates are defined as the number of first-time upper secondary graduates, regardless of their ages, divided by the population at the age at which students typically graduate from upper secondary schools in each country. In the two East Asian countries, Japan and Korea, graduation rates exceed 90% for both males and females, along with some other countries like Germany (and males in the Netherlands). On the other hand, the graduation rates in the U.S., Italy, Spain, and Sweden are at, or a little less than the average, of the 20 countries, while in Mexico only about 30% graduate from high school. Overall the figure indicates that the two East Asian countries do a better job than many countries in preparing young people to possess basic skills and knowledge necessary for further education or working life. Although in the figure Germany and the Netherlands also have high graduation rates, they have very different upper secondary programs than those in Japan or

Transition to Adulthood in Japan and Korea

53

Fig. 3. Percentage of Population that Has Attained at Least Upper Secondary Education by Age Group. Source: OECD (2001b).

Korea. In Germany and the Netherlands, both of which have strong apprenticeship programs, the majority of graduates complete pre-vocational/vocational programs rather than general programs: in 1999, about 75% of upper secondary students in these two countries were enrolled in pre-vocational/vocational programs. In Japan and Korea, however, we observe the opposite case. This reflects high demand for post-secondary education in the latter, which leads to a strong preference among students to choose general academic high school. Figure 3 shows the differences between the two age groups of 25–34 and 55–64 in the percentage of the population that has attained at least upper secondary education for selected countries in 1999 (OECD, 2001b). More than 90% of those aged 25–34 in Japan and Korea have completed upper secondary education. Only a few countries like the Czech Republic or Norway attained comparable levels of education. This high level of educational attainment in Japan and Korea is in sharp contrast to the two countries in southern Europe (Italy and Spain) where only half of the youngest cohort has completed upper secondary education. More impressive is the extent of change between the two age groups in Korea and Japan. In particular, in Korea the proportion of people who have at least upper secondary qualifications increased more than three times between the age groups

54

HYUNJOON PARK AND GARY D. SANDEFUR

Fig. 4. Percentage of Population that Has Attained at Least Tertiary Education by Age Group. Source: OECD (2001b).

of 55–64 and 25–34. Although the extent of increase is not as substantial as in Korea, a significantly higher proportion of young people completed at least high school in Japan compared to their older counterparts. We see significant increases in the proportion between the two age groups in other countries like Spain or Italy as well. However, young people in those countries have much lower levels of attainment than those in Japan or Korea. In Fig. 4, we see a dramatic increase in tertiary education in Korea and Japan. Only 14% of Japanese and 9% of Korean people aged 55–64 have at least a tertiary qualification, while for the age group of 25–34 the proportion is 45 and 35%, respectively. In fact, Japan has the second highest proportion of 25–34-year-olds who have attained at least tertiary education, following Canada (47%), among the total 29 OECD countries for which data are available in 1999.

Gender Differences in Educational Attainment It is remarkable to observe how gender gaps in educational attainment have reduced over time particularly in Korea. Specifically, the proportion of Korean men aged

Transition to Adulthood in Japan and Korea

55

55–64 in 1999 who have at least an upper secondary qualification is 43%, while the proportion for corresponding women is only 13%. However, among those aged 25–34, the gap between men and women is almost nothing. Although the gender gap in high school graduation was already small favoring men in the Japanese older cohort of 55–64, we can also observe noticeable improvement in women’s education. Now, a larger portion of younger women than men in Japan has at least a secondary qualification. A significant increase of educational attainment among Japanese and Korean women is evident at the level of tertiary education as well. The proportion of women who attained at least tertiary education increased more than four times from 10 to 46% between the two age groups of 55–64 and 25–34 in Japan. The expansion of higher education among Korean women is more dramatic: the difference in the proportion who completed at least tertiary education between the two age groups is about fifteen times with 2% for those aged 55–64 and 31% for those aged 25–34. A similarly dramatic expansion of higher education among women as well as men in Japan and Korea, however, did occur with different paths. While education expansion among Japanese women was mainly driven by their increase in attending 2-year junior colleges rather than 4-year universities, Korean women advanced relatively equally to junior colleges and universities (Brinton & Lee, 2001). In Japan, female students are over-represented at junior colleges, while male students are over-represented at universities (Ishida, 2002).

ECONOMIC ACTIVITY As one of the important events young people experience in their transition to adulthood, the transition from school to working life has received increasing attention from researchers and policy makers who try to understand the impacts of institutional characteristics on young people’s labor market situations (OECD, 2000; Shavit & M¨uller, 1998; Stern & Wagner, 1999). This growing interest in labor market conditions among young people reflects the declining economic status of young workers, represented by rising unemployment and reduced wages among young people, in many countries in the 1990s (Blanchflower & Freeman, 2000).

Labor Force Participation The change in labor force participation among young people in Japan and Korea displays different trends depending on their ages and gender. While labor force

56

HYUNJOON PARK AND GARY D. SANDEFUR

participation rates have been relatively stable among men aged 20–29, there has been a significant growth of labor force participation particularly among women aged 20–29 during the last few decades. However, the rate of labor force participation among those aged 15–19 for both men and women significantly declined, reflecting rising enrollment and the transition to tertiary education at this age group. Figures 5a and b present labor force participation rates across age groups in the two East Asian countries and the U.S. (ILO, 2003). In regard to labor market behaviors among young people, the figures show distinctively lower levels of economic activity of those aged 15–19 in Japan and Korea compared to American counterparts. Specifically, the labor force participation rates of men aged 15–19 are about 20% in Japan and only 9% in Korea, while the rate in the U.S. is more than 50%. Japanese and Korean women aged 15–19 have labor force participation rates lower than 20%, while about half of American women aged 15–19 participate in the labor force. Although the gap is narrower for those aged 20–24 than for those aged 15–19, the difference in the level of economic activity between American and the two East Asian young people is still substantial among men. But for those aged 25 or older, there is no significant difference in the extent of labor force participation among men in the three countries. Interestingly, women aged 20–24 in these three countries do not show significantly different degrees of labor force participation, though women aged 25–34 in Japan and Korea are much less likely to be in the labor force than are American women. Various factors might produce such a difference in the levels of economic activity between young people in Japan and Korea on the one hand, and the U.S. on the other, including different preferences for working among young people and variations in employment opportunities for young people. The preparation for university entrance examinations among high school students in East Asia might be particularly relevant for this different pattern of economic activity among those aged 15–19. Given a high level of high school enrollment and serious competition for entering top universities, high school education in Japan and Korea largely concentrates on preparation for entrance examinations and students spend a good deal of time preparing for these tests. Therefore, it is difficult for students to combine study with full-time or even part-time work. Another interesting pattern regarding the variation in labor force participation rates by age groups is that there still exists the typical M-shaped curve of female labor force participation in Japan and Korea. Substantial portions of women withdraw from labor markets around the time of marriage and childbearing, which results in a deep decline of labor force participation rates in the age groups of 25–34. Then the rates increase again as women who were out of

Transition to Adulthood in Japan and Korea

57

Fig. 5. Labor Force Participation Rates Across Age Groups (a) Men and (b) Women. Source: ILO (2003).

labor market during the periods of childbearing return to labor force after age 35. There is no evidence that this M-shaped pattern of labor force participation among Japanese women weakened until at least the mid-1980s (Brinton, 2001). In Korea, the M-shaped pattern still persists, though since the late 1990s there has been some

58

HYUNJOON PARK AND GARY D. SANDEFUR

change toward more female labor force participation at the ages of 25–29 (Kum, 2002). This M-shaped pattern of labor force participation across life course among Japanese and Korean women, in fact, is not apparent in Taiwan, which is another East Asian county. The rates of labor force participation among Taiwanese women increase to ages 25–29 and then continuously decline with age, resulting in a “single-peaked” pattern that is much close to the Western type. Brinton (2001) attributed the differences in the discontinuities of female labor force participation at the period of marriage and childbearing to the different nature of labor demand between Japan and Korea on the one hand and Taiwan on the other. Specifically, in contrast to Taiwan’s employment structure of small firms, economic systems of Japan and Korea on the basis of large firms provide a work environment less favorable for women to combine work and family roles. The relatively larger size of the public sector in Taiwan than in Japan or Korea also contributes to the continuous employment of Taiwanese women.

Unemployment Rate The unemployment rate has been widely considered as a key indicator of the difficulties that young people experience in the transition from school to work (OECD, 2000). Showing the overall rates of unemployment in the three age groups in 10 selected countries Figure 6a indicates significant cross-national differences in the degree of young people’s transition difficulties (ILO, 2003). If we focus on the age group of 20–24, the unemployment rate is about 6% in Japan and Korea, while it is about 10% in the U.S., Germany, Norway, and the U.K. Twenty percent of those aged 20–24 in Sweden and France are unemployed, and the unemployment rates in Spain and Italy exceed 30%. In most countries, unemployment rates decline with age, but the pattern of cross-national differences in the level of unemployment remains similar in the age group of 25–29. The figure suggests that in general young people in the two East Asian countries as well as Germany experience relatively little difficulty in their transition from school to working life, compared to those in Southern European countries like Spain or Italy. Recently, the unemployment rate has been criticized as not measuring well the difficulties young people experience in the transition from school to work (OECD, 2000). Because it measures the proportion of unemployed relative to those in labor force, the same unemployment rates may indicate very different situations depending on the degree of labor force participation among young people. When the labor force participation rate is high, the same unemployment rate indicates

Transition to Adulthood in Japan and Korea

59

Fig. 6. (a) Unemployment Rates by Age Groups (1996). (b) Unemployment to Population Ratio (1996). Source: ILO (2003).

that a larger proportion of people are unemployed among the total population of the age group. Thus, as an alternative, another indicator – the unemployment to population ratio for an age group – has been increasingly used in the literature.

60

HYUNJOON PARK AND GARY D. SANDEFUR

Figure 6b shows the ratio of the unemployed to the total population for each age group for the same 10 countries as in Fig. 6a. Because relatively larger proportions are in school among those aged 15–19 relative to those aged 20–24 or 25–29, the unemployment to population ratio is deeply reduced compared to the unemployment rate for the age group of 15–19 than for other groups. But the basic lesson from the figure is the same as with the indicator of the unemployment rate; overall, young people in Japan and Korea are less likely to suffer from unemployment than their counterparts in other regions. Unemployment among young people tends to be related with adult unemployment; a country with high adult unemployment is more likely to have high unemployment among young people (Stern & Wagner, 1999). For instance, Japan and Korea display relatively low youth unemployment rates (6.7 and 6.1%, respectively, among aged 15–24) and also relatively low levels of unemployment among those aged 25–54 (2.7% and 1.5%, respectively). In contrast, youth unemployment rates in Spain and Italy are more than 30% and unemployment rates among the ages of 25–54 are also relatively high with 19 and 9%, respectively. Therefore, the low levels of unemployment among young people often reflect the overall economy in a country. However, studies have also pointed out some distinctive linkages between high school and work in Japan that in part explain the low unemployment among young people. High schools in Japan are crucial in helping students find a job on the basis of long-term relationships with employers (Kariya, 1999; Rosenbaum & Kariya, 1989). Employers tend to recruit students recommended by specific high schools and this relationship between employers and schools persists through many years to become stable. Hence, this is an important mechanism through which the transition to work occurs in Japan and helps reduce unemployment among young people.

MARRIAGE AND CHILDBEARING Marriage A significant increase in educational attainment among East Asian people might affect marriage patterns and timing in this region. Figures 7a and b present the trends in the age at first marriage in Japan, Korea, and the U.S. Here the age at first marriage is represented by the singulate mean age of marriage (SMAM) for Japan and Korea, while it indicates the median age of first marriage for the U.S.6 What is evident from the figures is that in all three countries, the mean or median age at first marriage has increased overall for both men and women,

Transition to Adulthood in Japan and Korea

61

Fig. 7. Trends in Age at First Marriage (a) Men and (b) Women. Note: The age at first marriage is represented by the singulate mean age of marriage (SMAM) for Japan and Korea, while it indicates the median age at first marriage for the U.S. Source: UN (1990, 1997) and U.S. Census Bureau (2003b).

though it remained relatively stable during the 1960s and the first half of the 1970s in Japan and the U.S. Specifically, during the three decades from 1960 to 1990, the singulate mean age at first marriage increased from 25.0 years to 26.9 years among Japanese women and from 21.3 years to 25.4 years among Korean women. During the same period, the median age at first marriage among American women increased from 20.3 years to 23.9 years. Due to different measures of age at first marriage (singulate mean age or median age), it is difficult to draw a definite conclusion on cross-national differences. However, it seems likely that the age at first marriage among Japanese and Korean people is higher on average

62

HYUNJOON PARK AND GARY D. SANDEFUR

than that among Americans by roughly 2–3 years for both men and women (Inoue, 1998). To gain additional information on marriage patterns, we examine the proportions of never married men and women across age groups in the three countries (U.S. Census Bureau, 2003a, b). Let’s first focus on variations in the proportions of never-married women by age groups and countries in Fig. 8a. Up to the age group of 20–24, American women are more likely to marry than their counterparts in the two East Asian countries. For instance, 67% of American women aged 20–24 never married, while more than 80% of Japanese and Korean women in this age group never married. The proportion of never-married Korean women, however, declines steeply in the age group of 25–29 so that in this age group the proportion is higher for American women than for Korean women. Between the ages of 25–29 and 30–34, the decline is more moderate in the U.S. and hence there is no more substantial difference between American and Japanese women. In general, the figure indicates that the proportion of American women who marry before 25 years old is higher than those of Japanese and Korean women, while women in the two East Asian countries tend to marry in a more narrow range of ages than do American women. We observe similar patterns of cross-national differences among men, though the steep decline of the proportion of non-married Korean men occurs between the age groups of 25–29 and 30–34, which is 5 years later than for women.

Childbearing Students of the transition to adulthood have considered that early and out-ofwedlock childbearing can be a devastating experience in the lives of young people, especially young women (Sandefur, Eggerling-Boeck & Park, forthcoming). Having a child out of wedlock early in life can impede a young woman’s ability to pursue postsecondary education. It can also have a direct effect on her occupation and career by limiting her ability to work if she cannot find adequate and affordable childcare and an indirect effect through its effect on education. For this reason, policy makers are very concerned with both the level of out-of-wedlock childbearing and the factors that are associated with it. It has been widely documented that marriage and childbearing are closely connected and thus there is little non-marital childbearing in East Asian countries (Lin, Choe & Tsuya, 1999). For example, the percentage of births to unmarried women in Japan has been quite low around only 1% between 1980 and 1998, while it rose from 18 to 32.2% during the same period in the U.S. In fact, the percentage of births to unmarried women in the United States increased almost

Transition to Adulthood in Japan and Korea

63

Fig. 8. Proportions of Never-Married Across Age Groups (a) Women and (b) Men. Source: U.S. Census Bureau (2003a, b).

without interruption since 1940 with 3.8% until the middle of 1990s (Ventura & Bachrach, 2000). Although the comparable data of births to unmarried women are not available for Korea, we expect a very low level of out-of-wedlock childbearing in Korea similarly as in Japan.

64

HYUNJOON PARK AND GARY D. SANDEFUR

Fig. 9. Births to Women Under 20 Years Old per 1,000 15–19 Year-Olds. Source: UNICEF (2001).

Figure 9 presents the international comparable data on teenage birth rates for the 28 nations of the OECD (UNICEF, 2001). The number of births to women under 20 years old per 1,000 women aged 15–19 varies across countries from 3 and 4 births in Korea and Japan, respectively, to 52 births in the U.S. Again, here we confirm the significant difference in childbearing behaviors between American and East Asian young women.

LEAVING PARENTAL HOME Although many studies have compared the timing of leaving the parental home and factors associated with the decision to leave home for American and European young people (Corijn & Klijzing, 2001; Goldscheider & DaVanzo 1985; 1989), we know little about the pattern and timing of leaving the parental home among young people in East Asian countries. Although it deals with relatively older data, we exclusively rely on Yi et al. (1994) because it is the only study available regarding the issue of leaving home in East Asian countries. They provide census-based estimated net rates of leaving home in three East Asian countries (China, Japan, and Korea) and three Western societies (the U.S., France, and Sweden).

Transition to Adulthood in Japan and Korea

65

Fig. 10. Average Ages at which 25, 50, and 75% of Men and Women Leave their Parental Home. Note: East Asian countries used in this analysis include China, Japan, and Korea. Western countries include the U.S., France, and Sweden. Source: Yi et al. (1994).

Figure 10 presents the average ages at which 25, 50, and 75% of those men and women in the three East Asian countries (1970–1990) and the three Western countries (1968–1980) leave their parental home as estimated in the study of Yin et al. (1994). Although we should not neglect the differences within East Asian or Western societies, the similarities within each region are enough to focus our attention on the difference between the two regions. The figures indicate that in general young people in the U.S., Sweden, or France tend to leave their parental home earlier than those in China, Japan, or Korea. The average ages at which 25% of the men and women leave home are higher in the three East Asian countries than those in the three Western countries by 1.5 and 1.2 years, respectively. The differences increase across the quartiles so that there are 4.6-year and 3.9-year differences in the ages at which 75% of men and women leave their parental home. In short, from the figures we see that East Asian young people leave their parental home much later than their Western counterparts. In addition to strong cultural norms regarding intergenerational relationships that emphasize the responsibility of young people to take care of their parents, this pattern of the late age of leaving the parental home might be associated with overall lower levels of providing social welfare by states in the region. As we already saw, in Japan and Korea, individual families pay considerably higher proportions of the expenditures for tertiary education than in any other Western countries. In addition, financial aid from government for tertiary education like student loans and grants is not widely available (Park, 2004). Given the low level

66

HYUNJOON PARK AND GARY D. SANDEFUR

of governmental subsidies for education, students should rely on their parents for their education. Furthermore, many universities in Japan and Korea concentrate on cities with large populations. Hence, many students living in cities do not have to move to attend colleges, and in this context it is natural to live with their parents.

INTERRELATIONSHIPS AMONG TRANSITION MARKERS Although we examined the prevalence and timing of each marker of the transition to adulthood separately, it should be recognized that these various transition events are interrelated (Hogan & Aston, 1986). The trends in the length of time between transitions and the variation in the sequencing of transition markers have received particular attention among researchers to understand how the transition to adulthood has changed over time (Hogan, 1981; Modell, Furstenberg & Hershberg, 1976). Research has shown increasing individualization of life course events among American and European young people (Buchmann, 1989; Fussell, 2002b; Shanahan, 2000), including more variation in the sequence of transitions, dispersion in the timing of transitions, and the overlap of transitions. For example, studies suggest that a fairly orderly transition from full time schooling to work and then to marriage and childbearing is less common now than in the past (Hogan, 1981; Rindfuss, Swicegood & Rosenfeld, 1987). Many young people in America or Europe marry or have children prior to completing their education and intermingle periods of full-time work and full-time education. In contrast to increasing deviation from the traditional “standardized” sequencing of transitions observed among American and European young people, our description of each transition marker suggests relatively strong linkages among education, marriage, and parenthood in the process of becoming an adult in Japan and Korea. The emphasis on university entrance makes it difficult to combine work and study in high schools and it is rare for workers to return to education once they leave school for the labor market (Ishida, 1998). This “rigid structural barrier between the educational system and the labor market” is associated with a fairly orderly sequence of school-to-work transitions. Social norms about the proper ages and patterns of marriage and childbearing produce a strong connection between marriage and parenthood. As we have shown, births outside marriage are very rare in the two East Asian countries. Compared to their American counterparts, Japanese and Korean young people form families later in life and usually after finishing their education. Given the low level of welfare provision to families and children as well as little public financial support for higher education,

Transition to Adulthood in Japan and Korea

67

it is difficult for young people to combine parenthood and study in higher education. In short, in contrast to the relatively loose couplings among education, work, marriage, and parenthood in the American and European contexts, we observe the more traditional sequencing of the transitions from full time schooling to work and then to marriage and childbearing among young adults in Japan and Korea. The relatively strong connection among transition markers in the two East Asian countries remains at the same time that the age at which most individuals have attained all of these markers has increased. As we have seen, there has been a dramatic educational expansion and the age at first marriage has substantially increased for men and women resulting in later ages of family formation relative to those in America or some European countries. Even home leaving in these East Asian countries occurs much later than in America or European countries, with the exception of Southern Europe. In fact, this demographic pattern of a longer period of transition to adulthood, associated with educational expansion, increased female labor force participation, greater uncertainty in the labor market, the delay of marriage, and a longer stay in the parental home, is a common trend observed in many advanced industrial societies (Fussell, 2002b; Fussell & Gauthier, 2003). However, in the United States and some European countries, what we have found in the context of a prolonged transition to adulthood is a loose coupling between marriage and childbearing, including an increase in non-marital births and the rise of alternative pathways to family formation as seen in an increase in cohabitation (Buchmann, 1989; Fussell & Gauthier, 2003; Shanahan, 2000). It is interesting to see the extent to which currently existing standardized patterns of the transition to adulthood in the East Asian countries remain in the face of such global trends toward individualization of life course including greater variation in the sequencing of transitions and emerging new pathways to adulthood.

FAMILY BACKGROUND AND TRANSITION TO ADULTHOOD Differences in the patterns of the transition to adulthood are not only observed across cohorts but also within a cohort. Social class, race/ethnicity, and other social and demographic factors affect the sequencing and the timing of the transition to adulthood. Race or ethnicity does not play a substantial role in the homogenous Korean and Japanese societies. Recent studies have documented that various aspects of family background – socioeconomic status, cultural background, family structure, and so on – significantly affect important aspects of the transition to

68

HYUNJOON PARK AND GARY D. SANDEFUR

adulthood, including educational achievement (OECD, 2000) and educational attainment and childbearing (Sandefur, Eggerling-Boeck & Park, forthcoming). Better understanding the role that families play in young people’s transition to adulthood not only enriches our scientific understanding of this critical phase in individuals’ lives, but it also provides some guidance as to where social policy might be most effective in helping young people make a successful transition to adulthood. However, there is very little research examining how the effects of family background on the process of becoming adult vary across countries or regions. We have seen that in general young people in the three East Asian countries enter adult life with more education and that they marry and have children later. Childbearing and marriage are still strongly associated in East Asian countries. Furthermore, is there any reason for us to expect smaller differentials between young people from different backgrounds in their likelihood of making a successful transition to adulthood in this region than in other societies? Despite its importance for understanding cross-national differences in social inequality in regard to young people’s transition to adulthood, the lack of empirical research impedes our ability to provide reasonable answers to the question. Here we are only able to point out one interesting finding that might stimulate future research. As we introduced earlier, PISA 2000 is the international survey of reading, mathematical, and scientific literacy among 15-year-olds in 32 countries. In terms of the effect of the family’s socioeconomic status on student performance, the most interesting finding is that Japan and Korea show significantly less inequality of student performance by socioeconomic status as compared to the other countries in the survey. Specifically, in all countries, socioeconomic status is positively related to the student’s score on reading literacy. However, there is significant variation across countries in the extent to which socioeconomic status is associated with reading literacy. In particular, the slopes of the relationship between student scores on reading literacy and the composite measure of socioeconomic status are relatively flat in the two East Asian countries compared to other countries (OECD, 2001a), which indicates a weaker impact of family’s socioeconomic status on student performance.7 In short, the differences in reading literacy between students from higher and lower socioeconomic backgrounds are substantially smaller in Japan and Korea as compared to the other countries in PISA 2000. Although it is hard to draw any definite conclusion from this result, it does suggest substantial variation across countries in the extent of family background effects on educational achievement and the two East Asian countries show significantly less inequality in this dimension. At this moment, we are not in a position to explain this difference. More research is needed to examine whether

Transition to Adulthood in Japan and Korea

69

we may see a similar pattern of weaker impacts of family background for other transition markers in Japan and Korea.

SUMMARY AND CONCLUSIONS Our brief exploration of existing information about the transition to adulthood in Japan and Korea suggests that many features of the transition to adulthood in those societies differ considerably from what we know about the United States. Young people in the two countries attain high levels of education relative to other societies including those in the West. Family background does not seem to be as strongly related to educational achievement in East Asia as it is in the West. This is true even though East Asian governments seem to invest less public money in the educational system than do many Western societies. Families, on the other hand, invest considerable amounts of money in the education of their children. They also invest a good deal of time especially during the years of upper secondary education as students prepare for the very important college entrance examinations. The two East Asian societies are characterized by very different age-related patterns of female labor force participation from those in Western countries. The age profiles of labor force participation in Korea and Japan are M-shaped indicating that women enter the labor force upon completing their education, withdraw in large numbers during the childbearing years, and return to the labor force later in life. These patterns are consistent with a cultural tradition that prescribes childbearing and childcare norms for women and places less value on their activities in the labor market while their children are in primary and secondary school. Perhaps the most dramatic difference between East Asian countries and many Western countries involves family formation. Young people in East Asian countries leave home at later ages than those in the West, marry at later ages than those in the West, and have children at later ages than those in the West. Unlike the West where marriage and childbearing have become disconnected from one another as social processes, a much higher proportion of childbearing occurs in the context of marriage in East Asian countries than in the United States. The patterns of marriage and childbearing in East Asia resemble what many observers, including conservative policymakers, think should be the norm in the United States. Although the published data on Japan and Korea suggest many differences in the transition to adulthood from countries in the West, a good deal more research is needed to help us understand the nature and sources of these differences. Such research may also help us understand whether Western social policy, which hopes to promote a healthy transition to adulthood, can learn anything from some aspects of the East Asian experience.

70

HYUNJOON PARK AND GARY D. SANDEFUR

In this description, we pointed out differences between the two East Asian countries on the one hand and the American or European countries on the other. However, we are also aware of substantial differences in some aspects of the transition to adulthood between Japan and Korea. For example, the role of high schools for transition to labor markets is minimal in Korea, while the relatively strong linkage between high schools and employers is well established in Japan. We also described the different patterns of gender differences in post-secondary education between the two countries. Therefore, detailed examinations of both commonality and difference between the two East Asian countries will extend our understanding of variability in the patterns of the transition to adulthood and institutional factors resulting in such variation. Finally, when we described economic activity among young people, we intentionally limited our description to the period prior to 1998. The economic crisis that occurred in Korea in 1998 had huge impacts on labor market conditions for young people as well as adult workers. For instance, the unemployment rate increased two times among young people from 1997 to 1998: from 9.8 to 20.8% for those aged 15–19, from 7.1 to 14.8% for 20–24-year olds, and from 4.1 to 9.3% for those aged 25–29. Therefore, it was necessary to restrict out attention to the pre-crisis period. Some studies showed how economic change and historical events influenced the patterns of transition to adulthood for a specific cohort, resulting in significant variability across cohorts (Shanahan, 2000). It would be promising to examine how the economic crisis in 1998 has affected young people’s life course including family formation and leaving home as well as labor market situations, and how the impacts might vary by gender and family background.

NOTES 1. Although Esping-Andersen (1999) classifies the Japanese welfare system as close to the conservative welfare regime represented by Germany, there have been serious debates on whether welfare systems in the East Asian countries should be classified into the conservative welfare regime. The “Confucian welfare state,” or the “East Asian welfare model” is an alternative classification suggested by other researchers (see Cho, 2002). Here we use the term of “conservative” simply to indicate the low level of welfare provision without involvement in such debates. 2. For those dimensions on which data are available we provide information for selected representative countries of each welfare regime as classified by Esping-Andersen (1990): the U.S. and the U.K. for liberal welfare regimes, Germany for conservative regimes, Italy and Spain for Southern European types, and Sweden for social democratic regimes. 3. Specifically, those five measures are the percentage of children scoring below a fixed international benchmark determined for each survey of the followings: reading literacy of 15 year-olds in PISA 2000, mathematical and scientific literacy of 15 year-olds in PISA 2000, and mathematics and science achievement in the 8th grade. Children below these benchmarks are those who are at or below level 1 for reading literacy in PISA 2000, who

Transition to Adulthood in Japan and Korea

71

are below the 25th percentile among all children in OECD countries in PISA mathematical and scientific literacy, and who are below the median of all children in all countries in TIMSS 1999 mathematics and science achievement. 4. Specifically, the mean score on mathematical literacy among Korean students was 547 points, which was the second highest among 32 countries in PISA 2000. The male students’ mean score of mathematical literacy was 559, while the mean score of female students was 532 points. The difference of 27 points between Korean male and female students, along with the difference in Austria and Brazil, was the largest among countries in PISA 2000. Korean students scored highest for the scientific literacy with a mean of 552 points. But the gender difference of 19 points, favoring male students, was also the largest among the PISA countries (OECD, 2001a). 5. Japanese male students scored on average 561 and 547 points for mathematical and scientific literacy, while the average scores of Japanese female students were 553 and 554 points. The gender differences in both mathematical and scientific literacy were not statistically significant (OECD, 2001a). 6. The data are from UN publications (1990, 1997) for Japan and Korea, and U.S. Census Bureau, Annual Demographic Supplement to the March 2002 Current Population Survey, Current Population Reports, P20–547, “Children’s Living Arraignments and Characteristics: March 2002” and earlier reports for the United States (Internet release June 2003, http://www.census.gov/population/socdemo/hh-fam/tabMS-2.pdf). 7. The composite measure of socioeconomic status in PISA 2000 was constructed from a factor analysis on the basis of the following five variables: (1) parental occupation measured by the International Socio-Economic Index of Occupational Status (ISEI); (2) parental education; (3) family wealth indicating the material items possessed; (4) home educational resources; and (5) home possessions of “classical” culture (classical literature, books of poetry and works of art). This composite measure, thus, reflects an overall level of family’s economic, social, and cultural status. Linear regression analysis of reading literacy scores on the index of socioeconomic status was conducted for each country. Score point difference associated with one unit of the index of socioeconomic status was 21 points for Japan and Korea, while it was 48 points for the United States. The difference was even 60 points for Germany, while the average difference among OECD countries was 41 points.

REFERENCES Allmendinger, J. (1989). Educational system and labor market outcomes. European Sociological Review, 5, 231–250. Blanchflower, D. G. & Freeman, R. B. (Eds) (2000). Youth employment and joblessness in advanced countries. Chicago: University of Chicago Press. Blossfeld, H. P., & Huinink, J. (1991). Human capital investments or norms of role transition: How women’s schooling and carrier affect the process of family formation. American Journal of Sociology, 97, 143–168. Breen, R., & Buchmann, M. (2002). Institutional variation and the position of young people: A comparative perspective. Annals of the American Academy of Political and Social Science, 580, 288–305. Brinton, M. C. (2001). Married women’s labor in East Asian economies. In: M. C. Brinton (Ed.), Women’s Working Lives in East Asia (pp. 1–37). Stanford, CA: Stanford University Press.

72

HYUNJOON PARK AND GARY D. SANDEFUR

Brinton, M. C., & Lee, S. (2001). Women’s education and the labor market in Japan and South Korea. In: M. C. Brinton (Ed.), Women’s Working Lives in East Asia (pp. 125–150). Stanford, CA: Stanford University Press. Buchmann, M. (1989). The script of life in modern society: Entry into adulthood in a changing world. Chicago: University of Chicago Press. Cho, Y. (2002). Confucianism, conservatism, or liberalism? Examinations on welfare type in Korea. In: Y. Kim (Ed.), Debates on Characteristics of Welfare States in Korea (pp. 243–271). Seoul: Human and Welfare (in Korean). Cook, T. D., & Furstenberg, F. F., Jr. (2002). Explaining aspects of the transition to adulthood in Italy, Sweden, Germany, and the United States: A cross-disciplinary case synthesis approach. Annals of the American Academy of Political and Social Science, 580, 257–287. Corijn, M. & Klijzing, E. (Eds) (2001). Transitions to adulthood in Europe. Dordrecht: Kluwer Academic Publishers. Esping-Andersen, G. (1990). The three worlds of welfare capitalism. Cambridge: Polity Press. Esping-Andersen, G. (1999). Social foundations of postindustrial economies. Oxford University Press. Fussell, E. (2002a). The transition to adulthood in aging societies. Annals of the American Academy of Political and Social Science, 580, 216–239. Fussell, E. (2002b). Youth in aging societies. In: J. Mortimer & R. Larson (Eds), The Future of Adolescent Experiences: Societal Trends and the Transition to Adulthood (pp. 18–51). New York: Cambridge University Press. Fussel, E., & Gauthier, A. H. (2003). American women’s transition to adulthood in comparative perspective. Paper presented at the annual meeting of Population Association of America, Minneapolis, MN, May 1–3. Goldscheider, F. K., & DaVanzo, J. (1985). Living arrangements and transition to adulthood. Demography, 22, 545–563. Goldscheider, F. K., & DaVanzo, J. (1989). Pathways to independent living in early adulthood: Marriage, semiautonomy, and premarital residential independence. Demography, 26, 597–614. Hogan, D. (1978). The variable order of events in the life course. American Sociological Review, 43, 573–586. Hogan, D. (1981). Transitions and social change: The early lives of American men. New York: Academic Press. Hogan, D., & Astone, N. M. (1986). The transition to adulthood. Annual Review of Sociology, 12, 109–130. Inoue, S. (1998). Family formation in Japan, South Korea, and the United States: An overview. In: K. O. Mason, N. O. Tsuya & M. K. Choe (Eds), The Changing Family in Comparative Perspective: Asia and the United States (pp. 19–41). Honolulu, HI: East-West Center. International Labor Office (2003). Labor Statistics. Accessed March, 2003. http://laborsta.ilo.org/. Ishida, H. (1998). Educational credentials and labor-market outcomes in Japan. In: Y. Shavit & W. M¨uller (Eds), From School to Work (pp. 287–335). Oxford: Oxford University Press. Ishida, H. (2002). Educational expansion and inequality in access to education in Japan. Presented at the spring 2002 meeting of the Comparative Project on Higher Education, Prague, June 7–9, 2002. Kariya, T. (1999). Transition from school to work and career formation of Japanese High school students. In: S. Stern & D. A. Wagner (Eds), International Perspectives on the School-to-Work Transition (pp. 273–309). Cresskill, NJ: Hampton Press. Kum, J. (2002). Features and tasks of women’s labor markets. Working paper for Korean Labor Institute (in Korean).

Transition to Adulthood in Japan and Korea

73

Lin, H.-S., Choe, M. K., & Tsuya, N. O. (1999). Marriage in Taiwan, Japan, and South Korea. Presented at the annual meeting of the Population Association of America, New York, March 25–27, 1999. Marini, M. M. (1984). The order of events in the transition to adulthood. Sociology of Education, 57, 63–84. Modell, J., Furstenberg, F. F., Jr., & Hershberg, T. (1976). Social change and transitions to adulthood in historical perspective. Journal of Family History, 1, 7–31. Organization for Economic Co-operation and Development (1997). Education at a Glance. Paris: OECD. Organization for Economic Co-operation and Development (2000). From initial education to working life: Making transitions work. Paris: OECD. Organization for Economic Co-operation and Development (2001a). Knowledge and skills for life: First results from PISA 2000. Paris: OECD. Organization for Economic Co-operation and Development (2001b). Education at a glance. Paris: OECD. Park, H. (2004). Educational expansion and inequality in Korea. Research in Sociology of Education, 14, 33–58. Rindfuss, R., Swicegood, C. G., & Rosenfeld, R. (1987). Disorder in the life course: How common and does it matter? American Sociological Review, 52, 785–801. Rosenbaum, J. E., & Kariya, T. (1989). From high school to work: Market and institutional mechanisms in Japan. American Journal of Sociology, 94, 1334–1365. Sandefur, G. D., Eggerling-Boeck, J., & Park, H. (forthcoming). Off to a good start? Post-secondary education and the early transition to adulthood. In: R. A. Settersten, Jr., F. F. Furstenberg, Jr., & R. G. Rumbaut (Eds), On the Frontier of Adulthood: Theory, Research, and Public Policy. Chicago: University of Chicago Press. Shanahan, M. J. (2000). Pathways to adulthood in changing societies: Variability and mechanisms in life course perspective. Annual Review of Sociology, 26(9), 967–992. Shavit, Y. & M¨uller, M. (Eds) (1998). From school to work. Oxford: Oxford University Press. Stern, D., & Wagener, D. A. (1999). International perspectives on the school-to-work transition. Cresskill, NJ: Hampton Press. United Nations. (1990). Patterns of first marriage: Timing and prevalence. New York: UN. United Nations. (1997). Demographic yearbook 1995. New York: UN. United Nations Children’s Fund (2001). Teenage births in rich nations. Innocenti Report Card. No. 3. Florence: UNICEF Innocenti Research Centre. United Nations Children’s Fund (2003). Educational disadvantage in rich nations. Innocenti Report Card. No 4. Florence: UNICEF Innocenti Research Centre. United States Census Bureau (2003a). International data base. Accessed March 2003. http://www. census.gov/ipc/www/idbprint.html. United States Census Bureau (2003b). Annual demographic supplement to the March 2002 current population survey, current population reports, P20-547, “Children’s living arraignments and characteristics: March 2002” and earlier reports for the United States (Internet release June 2003. http://www.census.gov/population/socdemo/hh-fam/tabMS-2.pdf). Ventura, S. J., & Bachrach, C. A. (2000). Nonmarital childbearing in the United States, 1940-1999. National Vital Statistics Reports, 48(16). Hyattsville, MD: National Center for Health Statistics. Yi, Z., Coale, A., Choe, M. K., Zhiwu, L., & Li, L. (1994). Leaving the parental home: Censusbased estimates for China, Japan, South Korea, United States, France, and Sweden. Population Studies, 48, 65–80.

ETHNIC IDENTITY AND SEGMENTED ASSIMILATION AMONG SECOND-GENERATION CHINESE YOUTH Harry H. Hiller and Verna Chow INTRODUCTION The recent wave of immigration to North American society from new source countries challenges old theories of acculturation that were based on European immigration streams that assumed that ethnic retention was generationally conditioned. For Caucasian immigrants, it was assumed that assimilation was linear and that by the third generation, all traces of ethnic origin would be absent, save for a nostalgic interest in quaint and ephemeral aspects of an ethnic past labeled symbolic ethnicity (Child, 1943; Gans, 1979; Rumbaut, 1997; Waters, 1990). Since 1965 in the United States, and 1967 in Canada, changes in immigration policy suggest that alternative assimilation patterns may exist. Whereas previous immigration policy had discouraged non-Caucasian immigration, the new policy brought with it largescale immigration from Asia in particular which introduced a different element of race into assimilation expectations. For these new immigrants, race continues to be a marker whereby prejudice, stereotyping, and discrimination produce assumptions of “foreignness” regardless of generational status (Neckerman, Carter & Lee, 1999; Tuan, 1999).

Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 75–99 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10005-6

75

76

HARRY H. HILLER AND VERNA CHOW

In order to accommodate these changes, a new theoretical model has been articulated called segmented assimilation which assumes that there may be different modes of incorporation into the dominant society (Portes, 1997; Portes & Zhou, 1993; Zhou, 1997b). What this model has in common with older theories is a relationship between socio-economic status and assimilation. Whereas the older models assumed that assimilation occurred as a by-product of upward mobility, segmented assimilation suggested that blocked or downward mobility may be the new reality for some groups thereby retarding or resisting assimilation. A somewhat different emphasis though still rooted in status models is the social capital approach that focuses on how co-ethnicity supports vibrant communities that may either hinder mobility or support it (Zhou & Bankston, 1994, 1998). This paper proposes a shift away from mobility (or lack thereof) to that of identity as an indicator of segmented assimilation in relation to the new immigration. Instead of focusing on socio-economic factors, our goal is to examine socio-cultural factors as expressions of identity because identity is related to roles and social location in society (McCall & Simmons, 1978; Stryker, 1985). The new wave of immigration has produced a second generation of youth that provides a rich opportunity to examine whether segmentation exists and how members of this group understand themselves in interaction with others. The implication of segmented assimilation is that race can be a barrier to assimilation, so the question addressed here is how second generation Chinese understand their identity in relation to the dominant society and their familial socialization. How is this identity negotiated and how is it related to the process of segmentation? It will be shown that personal identity is divided into private and public spheres, and that the maintenance of ethnic identity in the private sphere (specifically the family and parent-child relations) allows for greater exploration in the public sphere that resists a segmented outcome for Chinese youth in Canada.

THEORETICAL BACKGROUND In segmented assimilation theory, it is proposed that there may be several trajectories to acculturation due to the fact that upward mobility is no longer a foregone conclusion. Phenotypic differences, combined with deindustrialization and the restructuring of the economy, have meant that some may experience the absence of mobility ladders, and youth in particular may be exposed to adversarial subcultures in inner cities that may support barriers to upward mobility. Portes and Zhou (1993) then propose that while integration into the white middle class may be the experience of some, others might experience permanent poverty and assimilate to an underclass culture. They also propose a third option whereby rapid economic

Ethnic Identity and Segmented Assimilation

77

advancement may occur but through the deliberate preservation of tight in-group solidarity and immigrant community values. Segmented assimilation then suggests that the straight-line conceptualization of assimilation through the generations needs modification in the light of a changing opportunity structure and different strategies of incorporation.1 At the very least, Gans (1992, p. 186) suggests that there may even be “delayed acculturation” when the second generation experiences a decline in mobility. Perlmann and Waldinger (1997) note that this mismatch between aspirations and shrinking opportunities may provoke a “second generation revolt” against unattainable dominant cultural values. Segmented assimilation theory makes an important point in noting that assimilation to the dominant culture is not necessarily an end result. The preservation of cultural distinctives may be forced on a group as a response to their low status or it may be a deliberate strategy in sectarian or enclave fashion to enhance full pursuit of economic advancement goals (Neckerman, Carter & Lee, 1999). This option is particularly relevant to more pluralist or multicultural models of society. While a more structuralist perspective puts the emphasis on the entrance status of the immigrant, it is also possible that for some groups ethnicity can be an asset or a form of social capital, which may assist in developing adaptation outcomes. The social capital of strong ethnic community supports or family values and structures may be absent in some instances contributing to “downward assimilation” or ghettoization while in other instances its presence may heighten self-esteem and foster academic and economic aspirations (Zhou, 1997a). For example, Zhou and Bankston’s study (1994, 1998) of a Vietnamese community found that a process of “selective Americanization” was adopted whereby obedience and filial obligation provided social controls that were highly instrumental in adaptive outcomes towards upward mobility. Waters (1994) also found that those of second generation Haitian or West Indian descent who saw more opportunities around them were more likely to be proud of their ethnic heritage than those who thought their opportunities were limited. Rumbaut’s research (1994) concluded that having immigrant parents who were high status professionals also led to greater identification with the immigrant culture and ethnic identity. In all of these cases, there is reason to believe that the straight-line theories of assimilation are open to review, for the new immigration and changing economy create the possibility of outcomes different from step-wise generational assimilation.

THE CANADIAN CONTEXT OF VISIBLE MINORITIES On the surface, segmented assimilation may have less utility when applied to Canada because of the absence of inner city ghettos typical of the American

78

HARRY H. HILLER AND VERNA CHOW

experience and because of the point system giving preference for admission to immigrants with advanced education and skills.2 Yet the term “visible minorities” is frequently used in Canada and is even specifically acknowledged in employment equity legislation (Henry & Tator, 1999, pp. 103–106). Henry (1994) coined the term “differential incorporation” to refer to the Caribbean-born community in Toronto where internal class divisions fragmented the group but in which children of working class parents often become an underclass of marginalized and alienated youth. While many Asian immigrants have taken employment as bluecollar workers, there is strong evidence that their children have been socialized to take advantage of mobility opportunities through advanced education. Hou and Balakrishnan (1996) found that with the exception of blacks, visible minorities have higher educational attainment than the average of the total population and the charter groups,3 and that their Canadian-born children were also considerably higher in educational attainment indicating that this was not just the result of selectivity in immigration. Yet these results do not necessarily translate into higher incomes or occupational levels. In short, assimilation models must be re-examined for the evidence is that there is not racial equality within Canadian society (Henry, Tator, Mattis & Rees, 1995). Since 1981, there has been a steady increase in the size of the visible minority population to about 11% of the Canadian population in 1996 (about three-quarters of which were Asian), and about 70% of this total was foreign born (Statistics Canada, 1998). The largest visible minority group in Canada is the Chinese. They made up about 27% of all visible minorities in 1996 and accounted for about 3% of the total Canadian population (Statistics Canada, 1998). Almost all visible minorities live in the largest urban centers. About 40% of all Chinese in Canada live in Toronto, 32% in Vancouver, and the rest in the other large cities. Approximately 74% of all Chinese living in Canada are foreign born. Of those born abroad, 29% immigrated before 1981, 32% immigrated between 1981 and 1990, and 40% immigrated between 1991 and 1996. Thus it is clear that Chinese are largely a recently migrated group and that this wave of migration has been continuing. Li (1998, p. 105) notes that this means that most native-born Chinese are under 16 years of age suggesting that a sizable cohort of second and third generation Chinese has been slow to develop. The percentage of Chinese persons with Chinese as mother tongue and language most often used at home is a significant indicator of a high percentage of foreign born in the population. Yet he notes that there is a language loss as those in subsequent generations begin to switch to English. The new wave of middle class immigration from Hong Kong and Taiwan beginning around the mid-80s brought a new image of Chinese from that of laborers to professional, white collar, and educated. Consequently, the proportion of Chinese in Canada with a university education is at least double that of other Canadians (Li, 1998, p. 121).

Ethnic Identity and Segmented Assimilation

79

A study of Chinese youth serves as an interesting test of segmented assimilation theory. The new wave of Chinese immigrants with high ambitions and greater skills, with social solidarity supported by a significant critical mass, and with a continuing supply of new immigrant co-ethnics raises the possibility that the third option described by Portes and Zhou of ethnic preservation as well as upward mobility might be possible. Race then may combine with cultural distinctives to support a form of segmented assimilation. One place to look for this outcome is in the second generation youth that must deal with the conflicts between the culture of their parents and the dominant culture surrounding them.

THE SECOND GENERATION: FROM MOBILITY TO IDENTITY The bipolar model that is frequently applied to the second generation suggests that the second generation is caught between the two worlds of the parents and the dominant culture, and that its members must negotiate in and out of the two segregated identities (Bacon, 1999). This type of “marginal man” analysis produces either a rather schizoid second generation or necessarily implies that the second generation will have a weaker ethnic identification than their parents. Another analytical perspective is to assume that the outside self is adaptive to the new culture while the inside self retains its ethnic core, or that the second generation learns to take the best of both worlds and synthesizes it into a new identity of cultural hybridity. Research on the second generation immigrant has not received much attention in Canada, and when it has, the focus has often been on mobility related issues such as income, occupation, and education (e.g. Boyd & Grieco, 1998; Breton et al., 1990; Isajiw, Sev’er & Driedger, 1993; Kalbach, Lanphier, Rhyne & Richmond, 1983). However, issues of identity and behavior (e.g. language use and friendship patterns) are also significant and reveal something of the way in which the second generation deals with their status. Tsang et al. (2003) studied how Chinese youth negotiate their identity but their specific focus was on “satellite children” where foreign born parents returned to the country of origin. Berry (1997) has developed a typology of acculturative attitudes that focused on the perceived importance of the maintenance of an ethnic identity and its cultural characteristics in relation to the importance given to maintaining relationships with other groups. Options varied from “assimilation” and “deculturation” to “separation,” or adopting traditions and values from two or more cultures labeled “integration.” Persons choosing a “Canadian” identity were more likely to experience deculturation, integration, or assimilation whereas those selecting an ethnic label were more likely to prefer

80

HARRY H. HILLER AND VERNA CHOW

separation (Tonks & Paranjpe, 1999). In contrast, those choosing an ethnicCanadian label were more likely to prefer integration. None of this work explicitly introduced race into the analysis but it does suggest that identity and race together might also be relevant to segmented assimilation theory. This matter is especially important given recent research in the United States on the emergence of what is called pan-ethnicity (Lopez & Espiritu, 1990). Building from the notion that ethnic identity is the result of a dialectical process involving not only your own self-identification but also how the host society identifies you (Nagel, 1994, p. 154), this process is especially applicable to racial minorities who are lumped together by the host society into categories as though they were homogeneous entities. For example, Portes and MacLeod (1996) have shown how American culture has created a new identity through the term “Hispanic” which lumps together Latin American minorities of diverse nationalities (e.g. Cubans, Columbians, Chileans) because of phenotypic similarities despite cultural differences. Kibria (1997) uncovers a pan-Asian American identity emerging among the Korean and Chinese second generation as a result of shared experiences and being racially labeled as Asian by the dominant society. While Kibria is careful not to assert that such pan-ethnicity is characteristic of all Asians, many Asians did perceive that their adherence to core values of Asian culture such as in family life and child-rearing, education, hard work, and respect for elders provided a behavioral basis for a pan-racial identity different from the dominant culture which might affect even matters such as dating and mate selection. Another issue indicative of assimilation patterns is ethnic language maintenance. Again, expectations in the past were that the first generation managed to learn enough English to survive economically, the second generation spoke the ethnic language at home whereas English was used in public life such as at school or work, and the third generation lost the ethnic language capability completely and adopted English as their mother tongue. But have these patterns changed? Portes and Schauffler (1994) found that among Spanish-speaking second generation immigrants in South Florida that there was rapid linguistic assimilation to English monolingualism. Speaking English at home and/or with friends increased the likelihood of the second generation assuming an American identity, and conversely speaking another language increased the likelihood of adopting an ethnic origin identity (Portes & MacLeod, 1996, p. 537; Rumbaut, 1994, p. 780). Yet Zhou and Bankston (1998) found that among Asians, fluent bilingualism among the second generation was much more likely than English monolingualism. This was primarily the case when they lived near co-ethnics, their parents lacked English proficiency, and repeated waves of immigration helped to keep active use of the language alive. Zhou and Bankston (1998) and Rumbaut (1994) conclude that ethnicity must be viewed not so much as a label, a tradition, or a national origin but a system of social relations in which parental relations and socialization, and kinship

Ethnic Identity and Segmented Assimilation

81

and friendship networks shape the nature and pace of assimilation. For example, the prevalence of three generation households among certain immigrant groups adds a different dimension in the presence of elderly persons that helps sustain ethnic traditions and norms (Perez, 1994). All of these observations suggest that language use is pivotal to ethnic identity, and that groups with active immigration streams and strong social ties are more likely to retain these ethnic identities. There is, then, considerable reason to expect that second generation Chinese in Canada may not fit straight line assimilation expectations – or at least that it is too early in the migration wave to assess its full impact. New Chinese immigrants continue to arrive, old world Chinese values and traditions may still be alive, and language patterns may be sustained through interaction with other Chinese in large urban communities where Chinese have settled. Chinese organizations and thriving Chinese commercial districts add a sense of robustness to the community (Lai, 1988), and, of course, there is also the question of whether a racial difference helps to sustain ethnic behavioral patterns. All of this is taking place in a society that champions the ideal of being multicultural (Li, 1999). When these observations are combined with the high achievement orientations of many Asians which has led to their designation as the “model minority” (Hurh & Kim, 1989), it is apparent that the applicability of segmented assimilation theory to second generation Chinese in Canada merits empirical testing. Are second generation Chinese in Canada moving towards assimilation or are there elements to their structural location within the society that prohibits, retards, or alters that response to the dominant society? More directly, how do the Chinese second generation negotiate that location in relation to matters of personal identity and behavior? It is widely recognized that the focus on immigration has usually been on the immigrants themselves and that the study of the second generation has been often ignored (Portes, 1996; Rumbaut, 1994; Zhou, 1997a, p. 91). This has largely been due to the fact that the second generation was viewed as transitional to full third generation assimilation. Now that source country origins of immigrants have changed and race has been added as a new dimension to the acculturation process, it might be helpful to look more closely at the second generation for help in understanding what we might expect in the future. Therefore this study focuses on second generation Chinese youth in Canada with specific emphasis on how they interpret and negotiate their identity and its behavioral components.4

THE STUDY DESIGN The large number of Chinese students in Canadian urban post-secondary institutions provides an opportunity to access young adults who are secondgeneration immigrants. Three post-secondary institutions were selected in the

82

HARRY H. HILLER AND VERNA CHOW

western Canadian city of Calgary, an urban center population of around one million with a Chinese population of approximately 60,000. While personal interviews may have also been a rewarding data generating mechanism for the study (compare Kibria, 2002), it was decided to use an open-ended questionnaire which respondents could complete at their leisure and without the influence of an interviewer.5 A purposive sampling frame was used as respondents were sought at places on these college campuses where large numbers of Chinese students gathered for study or socializing.6 The questionnaire was distributed by one of the authors (who is Chinese), which it was hoped would enhance the response rate. All students who completed the Consent Form and were handed a questionnaire also returned their questionnaire to a marked box to ensure anonymity. Obviously, there are many second generation Chinese who do not attend post-secondary educational institutions and the sample may be skewed towards those who have an upwardly mobile or middle class perspective. If indeed this is the case, the framing of identity issues may even be more germane because of their anticipation of direct participation in the dominant society. In order to participate in the study, both parents of the respondent had to be foreign born. The sample was composed of ninety respondents of which 53% were born in Canada, and 37% who had resided in Canada for more than ten years. Only 10% had resided in Canada for less than ten years. In spite of the 18–30 year old criterion in the sample design, observation of respondents led to the conclusion that most of them were between 18 and 22 having recently completed high school. This would suggest that most of the non-Canadian born came to Canada as small children, which means that most of their socialization took place in Canada. This group is either identified as the 1.5 generation or is considered second generation because they are the children of immigrants and have been educated here.7 Obviously the cut-point between the first and second generation is arbitrary, but if adolescence is made that dividing line, it is clear that virtually all respondents faced the acculturating pressures of the primary and secondary educational systems in Canada, and for the sake of this study can be considered second generation. Fifty of the ninety respondents were female and forty were male. The questionnaire was designed to encourage reflection with opportunities for the respondent to describe how they felt about the question. In that sense, the primary thrust of the research was the collection of qualitative data that would allow respondents to express themselves in their own words. The two lead questions, for example, were open-ended and were meant to allow the participant to discuss their own identity. An attempt was made to collapse these responses into appropriate categories for analytical purposes. The third and fourth questions sought to assess parental influence on the respondent as well as contrasting personal reflections

Ethnic Identity and Segmented Assimilation

83

about the importance of that heritage. The remaining open-ended questions dealt with attitudes towards social ties as friends, dates, and potential mates. In this way, primary socialization and peer socialization could be related to identity. Some questions with predetermined options were also included such as to rate their Chinese linguistic skills that are represented here in simple percentages. Overall, the size of the sample and the purpose of the study made the qualitative data the central focus.

THE FINDINGS Identity and Heritage Respondents were first asked how they described their identity (“How would you describe your identity?”). Following Portes and MacLeod (1996, p. 533) who also preferred a and-written self-designation over a fixed choice in determining ethnic identity, the goal was to capture the respondent’s own conception of their identity. Interestingly, being Chinese was a central identity for most respondents, but it was usually linked to the additional descriptor of being Canadian as well. The two dominant choices accounting for 66% of the respondents were Chinese-Canadian (41%) or Canadian-Chinese (25%). Only 7% saw themselves as just Canadian. 10% were uncertain or gave no response at all. But 17% thought of themselves in racial/ethnic terms as Chinese, Hong Kong Chinese, or Oriental. The fact that “Chinese” and “Canadian” were linked together is perhaps no surprise for the second generation, but the fact that two different word ordering appeared suggested that ordering may be of special significance. While putting the racial/ethnic label first when linked with Canadian (e.g. Chinese-Canadian) is a typical Canadian pattern, informal feedback was received from some respondents that the first term indicated a primary identity, and that is why both orderings appeared.8 If this is so, then those who identify themselves as “Chinese-Canadian” (by far the largest group) may have a special meaning. Putting Chinese first (as in Chinese-Canadian, or Chinese-born Canadian often abbreviated as a “CBC”) might indicate a primary Chinese identity in the Canadian context. When combining these respondents with this interpretation of Chinese primacy with the other more racial/ethnic responses of Chinese, Hong Kong Chinese, and Oriental, the result is that almost 60% of the respondents viewed themselves in racial/ethnic terms. Another interpretation is that the first word in a hyphenated naming is the adjective and that the second word is the important one. As noted earlier, “ChineseCanadian” (the classic hyphenated Canadian who views the hyphen as the heritage designator but “Canadian” as the core identity) was the largest single group but it was not a majority. Similarly, a “Canadian-Chinese” would be someone who is

84

HARRY H. HILLER AND VERNA CHOW

primarily Chinese, and Canadian is just an adjective. Many in this category actually used the term “Canadian-born Chinese” (Note that the CBC abbreviation works again). While it is impossible to say which interpretation is correct, it is likely that order conveys some form of identity preference for some people since both orders were used.9 But it is clear that whichever way the data are interpreted, the majority see themselves as a blend of two identities with enduring qualities understood in racial/ethnic categories. In other words, it was our perception that linking the label Chinese with Canadian in whatever order did have meaning because the term Chinese had not just ethnic meanings but racial meanings as well, and that there was a clear sense of a unique status within the society. At the same time, few understood their identity in pan-ethnic or national terms and always linked being Chinese with being Canadian. When length of residence is correlated with ethnic identity, it was not only those who had been here for less than ten years who selected Chinese, Hong Kong Chinese, or Oriental, but such choices were also made by those who were born here or lived here for longer than ten years. The fact that respondents would still utilize such explicit racialized terms is significant. Those who were born in Canada were only slightly more likely to respond that they were Canadian and they were clearly more likely to choose Canadian and Chinese. What is important is that while persons born in Canada were somewhat less likely to have a racial/ethnic identity, it is clear that the majority of persons in all residential categories identified themselves in racial/ethnic terms, especially if the ordering of the hyphenated identity had special meaning.10 It is very clear that race is an important basis for that identity because of its visibility in spite of how the respondent feels inside. In relation to the dominant society, it is something that cannot be denied. I think of myself as a banana – white on the inside and yellow on the outside . . . I’m Chinese but I’m Canadian. I have slanted eyes. I have to accept that I’m Chinese, but in the inside I still have a good laugh at the Chinese customs sometimes. Being that I came here when I was about two months old, I think that I’ve always felt like everyone else and not Chinese. But I guess I would describe myself as Chinese.

Public socialization within the dominant institutions of the society helps the second generation to feel that they belong, and yet a racial difference from the dominant group requires that their identity be at least partially modified. On the other hand, their private socialization at home is rooted in Chinese cultural traditions and provides the native tongue, which creates a split between how they see themselves at home and how they see themselves outside the home.

Ethnic Identity and Segmented Assimilation

85

I feel Chinese at home but Caucasian outside my home. I find there is a lot of conflict between my Chinese heritage and my Canadian identity.

The dual spheres then between private (home) and public (outside the home) are also countered by the struggle between a cultural definition of Chinese and a racial one. I describe myself as Chinese Canadian because I consider my Chinese values, tradition, and culture first. I am Chinese and I think it is very important not to think that I am not. I think that my ethnic identity could be stronger because I feel it weakens generation after generation.

Here identity is given a very clear ethnic/cultural spin because it is either a priority or it is assumed that it is undergoing a transformation and loss over time. The second open-ended question asked “How do you feel about your Chinese heritage?” The question was meant to deliberately be vague about what “heritage” included but aimed to ascertain generally how they felt about their Chinese identity.11 Perhaps most surprising was the fact that independently, over 50% of all respondents actually used the same word “proud” to describe their feelings about their heritage. They used words such as, “proud to be who I am because it adds another dimension to who I am,” “proud because it defines who I am,” “proud because of the culture, art, and moral teachings,” “proud, and enjoy participating in celebrations and learning about it from my parents,” and, as one respondent put it, “I am proud of it now, but when I was younger I just wanted to fit in. This usually meant ignoring my heritage.” When these responses are added to the others of a positive nature (almost 30%) such as “good,” “important,” “accept and respect it,” “very interested in it,” and “strong feelings towards maintaining my heritage,” 80% of all respondents can be accounted for. The remainder were either “neutral,” “depends,” “confusing,” “don’t care,” or no response. Thus, in general terms, there was overwhelming positive feeling towards their Chinese heritage. In spite of some dilemmas pertaining to racial distinctiveness and cultural retention, and the split between their private/home identity and public identity, there was an overwhelming positive affirmation of their identity as Chinese.

INTERACTION AT HOME A key element in the preservation of an identity is the cultural artifacts and customs that help to sustain that identity. Parents are the ones to introduce their children to those traditions, and it is important to know whether there are any generational

86

HARRY H. HILLER AND VERNA CHOW

differences in attitudes towards them. It could also be argued that these customs stress more the ethnic/cultural side of being Chinese. When respondents were asked how important Chinese customs and culture were to their parents, it is somewhat surprising that only 34% said “Very Important,” and 64% said “Somewhat Important.” Only 2% said “Not Important.” Youth who had been here less than ten years were more likely to say “Very Important” although not exclusively so. Respondent’s justifications for their answer seemed to be very similar regardless of which option they chose in that repeated references were made to cultural artifacts such as “eating rice,” “following the lunar calendar,” “celebrate special days such as Chinese New Year,” “practice some Chinese medicine such as herbs and pills,” or “go out for dim sum.” What seems to differentiate those who say it is “Very Important” is a deeper recognition of the religious aspects and filial aspects of Chinese culture. For example, ancestor worship, Buddhist practices, the “plate of rotting fruit,”12 or choosing Chinese entertainment are mentioned. While a few pointed out that their parents were eager to maintain Chinese culture in some detail, the more common response was that parents were concerned not to put too much pressure on their children at the same time that they instilled a sense of appreciation for at least fragments of Chinese culture. For example: My parents take the issue of respect very strongly, but are willing to try out new things. They believe that they are in a different place and different time now, so life is adjusted accordingly, that not everything is maintained or followed. The culture is interwoven into their decisions, with respect to the older generation, and filial piety is still considered important, however, their lifestyle is not ruled by many of the Chinese customs.

In short, the data suggests that there is not the dogmatism and rigidity that sometimes is found amongst the first generation where adherence to traditional patterns of behavior produces generational conflict. Interestingly enough, when respondents were asked about how important Chinese customs and culture were to them personally, their response mirrored that of their assessment of their parents with 73% choosing “Somewhat Important” (compared to 64% for their parents). A few more said “Not Important” (10%) and a few less said “Very Important” (16%). There was usually very little difference in the explanations between “Very Important” and “Somewhat Important” as they all seemed to stress the need or desire to preserve their heritage and family history, and that this became more clear to them as they matured. As I grew older, I feel more and more regretful in losing some of my culture, that recently I try to recover what I’ve lost, for example, speaking Cantonese or attending more Chinese events. I want to retain the customs and culture and pass it on to my children.

Ethnic Identity and Segmented Assimilation

87

As I grew older, I became more aware of my heritage, and as of now I am very interested in my Chinese heritage but I would still consider myself Canadian. It is important to keep some culture and customs to know them, and it is always useful to know so that you know your background. I believe if I don’t continue to participate in Chinese customs, I will forget and no one will teach my children and I believe strongly in family values and loyalty Chinese families have. Since it is difficult to be Canadian and Chinese at the same time, it helps to learn about our own culture and to gain an understanding of it.

But there are many others who are also somewhat ambivalent about Chinese culture recognizing that it has its place but that personally there is considerable distance from that culture and ignorance of it. It’s only important in that it is fun to follow the customs but I don’t know the meaning behind the customs. I am only aware of what customs and traditions my parents pass on, but many of them I am not sure about and I try to go by the Chinese way, but realistically there is little support to do so. It doesn’t matter, I just need to know a bit and that is good enough for me. I don’t really know much but maybe someday I’ll want to learn about them. I am aware of them but unless it is something important, I only follow the customs at my convenience.

On balance, there is a clear leakage away from intimate knowledge of Chinese culture, but this loss is counterbalanced to a considerable degree by a desire for some heritage retention. Second, while it might be possible to speculate about whether these second generation Chinese might be underestimating the importance of Chinese culture to either themselves or their parents, it does seem to be clear that few see embracing this ethnic culture as a type of counter-culture to be preserved in the midst of an alien dominant culture (as segmented assimilation theory would suggest).

SOCIAL CHOICES: FRIENDSHIP, DATING AND MARRIAGE SELECTION Friendship, dating, and intermarriage are three significant indicators of in-group solidarity and three open-ended questions were asked about these matters. The first question aimed not to determine whether the respondent had Chinese friends but how important those friendships were. In response to the question “How important

88

HARRY H. HILLER AND VERNA CHOW

to you are friends who are Chinese?”, slightly over half (57%) said something to the effect that it was important. A sense of being in similar circumstances led to statements like “it is easier to relate to them,” “they understand what it is like to be Chinese,” “we share the same experience,” “because they bring security and comfort,” or “it is easier to identify with someone with the same background.” This commonality was based on race but also included shared experiences and similar home backgrounds. On the other hand, about 40% deliberately rejected that way of thinking that friendship based on race/ethnicity should serve as a basis for interaction. The declarative nature of these feelings were conveyed with strong statements such as “race shouldn’t be important,” “many of my friends come from different backgrounds,” “friendship goes beyond race,” or “all of my friends are important regardless of race.” While many respondents understood the value of ingroup relations, there appears to be a clear rejection of racial exclusion in friendship formation, even to the point that some are almost militant about being more open. In one way, having some Chinese friends in a pantheon of friends could be considered a rather inclusive ideal since the question does not ask whether a person should have “only” or “mostly” Chinese friends. A better test may be dating and marriage partner preferences. Quite unequivocally, 82% said they would “consider dating a non-Chinese” and a slightly less 75% said they would “consider marrying a non-Chinese.”13 Using the word “consider” may have affected the responses as it is somewhat more tentative than actually doing so. Nevertheless the fact that there was overwhelming openness not only to inter-group dating but exogamy in marriage choices as well suggests the lack of strong in-group ties and even the downgrading of such ties into the future. Those who preferred a Chinese partner in dating and marriage provided similar kind of rationales such as commonality in values and family backgrounds or the understanding and preservation of Chinese customs. Some had tried dating non-Chinese and it did not work well. But the overwhelming response was that racial/ethnic factors were not as important as personal compatibility and romantic love. “As long as it is the right person, it does not matter what race they are,” “no matter what, marriage is only for love,” “if they are willing to accept me for who I am,” “what is important is what is in the inside of the person,” “in Canada boundaries should not be important,” or “everyone is equal.” Whether these were platitudes for equality as opposed to ultimate bases for action of course might be debatable. Many respondents acknowledged that marrying someone who is Chinese might have advantages but they did not want their choices restricted by that fact alone. In fact, some respondents referred to strong physical attraction to members of other racial groups. In retrospect, using the word “consider” might have biased the response as it is widely recognized that in-group marriage is often thought to be at least a preference among many second generation Chinese.

Ethnic Identity and Segmented Assimilation

89

LANGUAGE USE The evidence that has been presented above of considerable openness to the dominant society finds an interesting counterpoint in language use. It is important to know that this second generation was overwhelmingly (83%) taught Chinese by their parents as their first language and only 13% were taught English as their first language. Ninety-one percent continue to speak Chinese with their parents with 52% “almost always” doing so and 39% “sometimes” doing so. In other words, the use of the Chinese language is alive and well with this second generation in its spoken form. Thirty-one percent claim to speak Chinese “very well” and 48% claim to speak it “reasonably well.” Eighteen percent know the language but speak it “poorly” but only 3% of all respondents say that they do not speak it at all. However, reading and writing Chinese is a totally different matter with 31% claiming that they cannot read Chinese at all and 36% being unable to write Chinese. A further 48% say they read Chinese poorly and 43% say they write it poorly. As expected, those who have been here less than ten years were more likely to be able to read and write Chinese very well whereas those who were born here or resident here more than ten years were more likely to be those who had no ability to read or write Chinese at all. Overall, however, the data indicate that the spoken form of the language is critical to communication at home, and even the written form of the language survives in some rudimentary fashion among second generation Chinese. Whether it can do so beyond this generation, however, is a major question. Yet to this point, speaking Chinese strengthens home ties and provides oral familiarity with others in the Chinese community. In fact, oral use and familiarity with the Chinese language appears to be the most important behavioral indicator of being part of that community.

DISCUSSION Perhaps the most important finding of this study is that these second generation Chinese possess a very limited sense of segmentation from the dominant society, and that this result is a consequence of a clear differentiation between the public and private sphere. While they are aware of the racial difference from the Caucasian majority, and while some articulated that it had a social-psychological impact on them, they fought hard to overcome any disadvantages which race might produce. They resisted in-group exclusivity at the same time that most seemed comfortable with their own identity between the dominant society and the culture of their parents. Rather than belaboring intergenerational conflicts or rebelling against their background, most seemed quite comfortable with who they were. The question is,

90

HARRY H. HILLER AND VERNA CHOW

why is this so? Five factors showing how the self is affected by interaction with others can be identified. The most surprising result was that the response to an open-ended question with no prompts or prescribed choices led to a majority of the respondents independently selecting the word “proud” to describe how they felt about their heritage. The strength of this positive affirmation not only reveals something about their inner selves (e.g. tranquility and lack of anger or confusion) but the way that they have been socialized. In-group pride might be a strategy for dealing with racism but is also a form of race socialization that occurs within families (Kibria, 2002, p. 41; Lyman & Douglass, 1977). For these second generation Chinese, race may be acknowledged as a limitation but was not perceived as a disadvantage that should be jettisoned. Instead, group pride taught at home and fostered within the ethnic community was accepted with confidence. Furthermore, pride in the heritage was a strong positive emotion that empowered rather than disabled and suggested that roots were to be affirmed and not rejected. Such pride may also be linked to a high achievement orientation and the “model minority” concept (Hurh & Kim, 1989). Again, this picture of a second generation is very different from that presented in the earlier literature on the children of European immigrants who sought to distance themselves from the culture of their parents, and may also be quite different from the experience of other racial groups who were made to feel inferior in a white-dominant society.14 It also is in contrast to the social psychological effects described by Lyman (1977, pp. 18, 19) of marginality, anomie, and selfestrangement experienced by children in the earlier phase of Chinese migration (see also Chan, 1998, p. 153). The second interaction factor of importance pertains to the social context of language practices. Portes and Rumbaut (2002, p. 144) use the term dissonant acculturation to refer to the problems that result when children become English monolingual and their immigrant parents are not fluent in English. When the parents are bilingual and/or when the children are bilingual, they refer to the end process as consonant acculturation. They argue that if the second generation is fluently bilingual, there is less likely to be loss of self-esteem or shame about the culture of the parents. In our study, while we did not ask questions about parental language use, it was clear that almost all of this second generation were bilingual which facilitated comfortable interaction in both the private sphere of the home and in public. While their use of the Chinese language in its written form was questionable, what was important is that they were able to communicate with parents, grandparents, and family friends in social contexts requiring the language and facilitating a sense of belonging.15 Instead of feeling marginalized by either their own culture or the dominant culture, they could interact comfortably in both.

Ethnic Identity and Segmented Assimilation

91

Third, the attitudes of parents seems also to play a significant role for our second generation respondents in relation to the issue of generational alienation. Our data suggested that the second generation clearly understands that their parents are also either fluently bilingual or are open to bilingualism in spite of the fact that Chinese is the home language. One of the reasons for this might be the fact that some Chinese immigrants came from places like Hong Kong where English language use is common. Given the fact that many of these new wave immigrant parents did not enter the country as laborers but as professionals, Chinese may be the language of the home, but many of them are in the labor force and use English outside the home as well. Thus the second generation have more in common with the first generation in that they both live in two linguistic worlds that reduce the dissonance with the home environment. Furthermore, the majority of respondents indicated that for both them (73%) and their parents (64%), retaining Chinese customs and culture was only “Somewhat Important.” This suggests that most understand their parents as being open to change and flexibility themselves with regards to traditional customs and practices, and that they acknowledged their own need to adapt to their new environment (compare Chan, 1998, pp. 128, 129).16 Following Zhou and Bankston’s study of Vietnamese youth in New Orleans (1994, 1998), it appears that this racial/ethnic identity as structured through family socialization is more a resource in acculturation and achievement than a disadvantage.17 Fourth, the policy underpinnings of a multicultural society legitimates the persistence of ethnic differences in the private sphere, and to an extent also in the public sphere.18 The statements of many respondents suggested that racial differences had often been transformed in their minds to ethnic differences. To be a “banana” (yellow outside and white inside) suggested a keen awareness of feeling inwardly like others in the dominant society at the same time that some customs and language use could be different. To an extent then, racial differences were reinterpreted as ethnic differences. The ideology of the Canadian multicultural state legitimates cultural retention (usually fragments of culture) and minimizes racism by stressing the naturalness of clinging to cultural groups, even when bounded by race. Yet if whiteness is the reference point for the Canadian identity, then non-whites are “othered” and non-meltable in important respects (Kelly, 1998). It is unclear how the second generation will handle the issue of race but it is clear from the data that the dominant culture legitimates the ethnicization of their racial identity, and many respondents demonstrated a keen interest in either retaining or modifying their ethnic heritage, a practice supported by the ideology of a multicultural society (Roberts & Clifton, 1982). Fifth, it has often been noted that identities may be affected by the experience of prejudice and discrimination. In their panel study in 1992 and 1995, Portes

92

HARRY H. HILLER AND VERNA CHOW

and Rumbaut (2002, pp. 155–181) were surprised to find a significant decline in the proportion of their second generation respondents who identified themselves as hyphenated Americans, preferring instead a foreign identity or pan-ethnic identity. The Chinese, for example, showed increases in a pan-ethnic identity. Portes and Rumbaut link this change to external circumstances such as the influence of Proposition 187 in the state of California or attending an inner city school where disadvantage and a gang culture may exist. This identity shift is understood as a reactive formation whereby a defensive identity develops in relation to the dominant group and other minorities. The experience of second generation Chinese living in large Canadian cities now involves interaction with large numbers of co-ethnics. While the presence of Chinatowns serve as active retail and entertainment centers (Lai, 1988), the Chinese population tends to be dispersed throughout the urban area without significant concentration. There have been few high profile events in Canada in recent years which have galvanized the Chinese population as outsiders or as under attack by nativists. Consequently, no defensive identities have arisen. Some of the second generation have clearly had experiences that serve as reminders of racial differences. Without asking any explicit questions about discrimination or prejudice, some respondents volunteered such information. As a child, I felt out of place due to discrimination against Chinese kids as there were few of us, but now I am proud to be Chinese and sometimes I wish I knew more about Chinese culture.

These kinds of observations suggest that forms of disadvantage may produce some alliance with a pan-ethnic identity in spite of differences in origin and background among Chinese people but there is little sense that this discrimination is debilitating or leads to attempts to repudiate the parental identity. The relationship between phenotypic traits and the social construction of race also suggests that interaction with others also impacts identity. In the American context, Tuan (1999) refers to this as the “authenticity dilemma” of being considered neither real Americans nor real Asians, which encourages the identification with hyphenated spaces. Because of prejudice, discrimination, and stereotyping, assumptions of foreignness may persist regardless of generational status. Tuan’s study of persons of Chinese and Japanese ancestry suggests that the commonalities of experiences by non-whites may push some into retaining a racial consciousness. The findings of this study seem to suggest that dual forces are at work, the acceptance of assimilation through public identity but also of ethnic preservation particularly in the private sphere. And throughout this process, ascriptive characteristics (race) serves as a shadow rather than a source of segmentation. In succeeding generations, if culture tends to erode, one can only guess about how

Ethnic Identity and Segmented Assimilation

93

race might affect identity, particularly as the position of the dominant Caucasian group erodes and Canadian society becomes more multi-racial.

CONCLUSION It is obvious that the expansion of the Chinese community in Canada in recent years and the escape from the lower social class position of earlier waves of Chinese immigration has transformed the structural location and identity of Chinese persons as individuals and as a group. Consequently, parents are more flexible and more facilitative (rather than resistant) to the construction of their children’s own identities in their new society. Perhaps they also experience less disadvantage, marginalization, and discrimination than earlier waves of Chinese immigrants (Lyman, 1977, 1986). In any case, segmentation as assessed by identity and sociocultural resources appears to be considerably tempered by a willingness to adjust and adapt to a multicultural society that legitimizes ethnic culture particularly in the private sphere of the family. The distinction between the private sphere of the family where language and customs are retained are in direct contrast to the public sphere where assimilation as measured by open attitudes towards friendship selection and mate selection are proceeding rapidly.19 Whether indeed these attitudes of young adults represent idealistic thinking or unrealistic optimism as opposed to actual reality may be an important point but, for example, even if ideas about mate selection are more ideal than real, they do reflect an attitude about assimilation that many must think are attainable and desirable. Has the segmented assimilation theory been confirmed among second generation Chinese in Canada? Recall that the expectation was that race and a changing economy and labour market shifts were introducing different paths to integration in American society including blocked mobility experienced by an ethnic/racial urban underclass as well as rapid economic advancement through the preservation of tight ethnic cultures. This study has no data on socio-economic mobility but it does assess the evidence for socio-cultural segmentation from the perspective of interactional identities among potentially upwardly mobile post-secondary students. The evidence presented here is that second generation Chinese do not view themselves as part of a tightly-knit ethnic culture that participates in only selective acculturation (Portes, 1997, p. 815) to the dominant Canadian culture thereby creating a defensive identity. In that sense, segmented assimilation theory must be rejected. On the other hand, race/ethnic identities, particularly in the private sphere, appear to provide an anchor that supports assimilation in the public sphere rather than serving as a hindrance to it. Rather than the result being a segregated

94

HARRY H. HILLER AND VERNA CHOW

identity typical of the second generation in other contexts, the result is more of a situational identity primarily as a consequence of perceptions of the attitudes of their parents. While it may appear that assimilation is proceeding in a similar manner to white immigrants, race combines with ethnicity to support the existence of a real boundary that appears permeable rather than a barrier. In other words, there remains some evidence that assimilation trends are counterbalanced by race/ethnicity, which might combine to structure associational preferences at least to some extent, but that the boundaries to greater integration are not impermeable.20 As one respondent put it, It is quite confusing sometimes to be Chinese in a Canadian society, but also difficult to be Canadian with a Chinese look.

Nevertheless, experiences of racial exclusion and discrimination have not been severe enough among recent Chinese immigrants to Canada to produce a strong identity built on race (Kibria, 1997). These conclusions seem to support similar findings in the United States that Asians do not conform to the stereotype of other racial minorities in that they have higher incomes and education, high rates of intermarriage, and less residential segregation (Lee, 1998). One of the key themes of this paper is that parent-child relations are an important ingredient in the development of ethnic identity as one might expect when discussing generational change. This focus serves as a significant addition to the literature on segmented assimilation by replacing mobility with identity, and noting how in a pluralist society identity switching (Lyman, 1977, p. 207) occurs from the private sphere of the home to the public sphere and vice versa, and that this switching occurs apparently with the support and participation of parents themselves.21 Ethnic identity is indeed fluid, situational, dynamic, and volitional for the second generation. For the Chinese second generation, it involves both choice (agency), and ascription (structure), and situation (public vs. private), and is the result of a dialectical process involving insiders (other Chinese persons) and outsiders (non-Chinese, and especially Caucasians) (Nagel, 1994). Identity is not so much related to a common origin as it is the result of negotiation and situation differentiating the public and private spheres (Yancey, Erickson & Juliani, 1976). To the extent that Canadian society is defined in racial terms by the Caucasian majority, the Chinese second generation seems to be aware of its minority status at the same time that it resists segmentation and seeks accommodation and incorporation. Recognition of being different from the dominant group has not produced reaction formation. Evidence from other ethnic/racial groups or from persons less likely to be of middle class status may produce different results. But from this ethnic/racial group at this stage of the life cycle, and among the second generation of this immigration wave in the Canadian context, it appears that

Ethnic Identity and Segmented Assimilation

95

segmentation is minimal. At the same time, ethnic preservation at home provides an identity anchor that encourages full participation in the public realm.

NOTES 1. It is interesting to note that Park (1950, pp. 194, 195) himself eventually reconsidered his race relations cycle and acknowledged that instead of assimilation some groups could become permanently institutionalized with a minority status. See also Gordon (1964, p. 235) who after emphasizing assimilation acknowledges conditions of structural separation. 2. Both the United States and Canada have experienced almost a complete shift away from the original source countries of Europe so that non-Caucasians are by far the largest group of immigrants. Both countries also have similar assimilation issues with the aboriginal population (Ponting & Kieley, 1997). Differences do exist though in the greater predominance of race in the United States as represented by a larger resident black population and a significant cross-border Mexican migration as well as migration from Latin America. 3. The term charter group is a term used in Canada to describe the two ethnic groups (British and French) who were the first to immigrate and settle in the country (other than aboriginals) and who controlled the nation-building process. In this regard, the British are considered the higher charter group because they had more control than the French. 4. It is significant that one important study of ethnic identity in Canada (Breton et al., 1990) did not include the Chinese in their study because the second generation was too young. Most of the data for that study was gathered in the late 1970s. 5. One of the limitations of this procedure was that it eliminated any opportunity for clarification of the information the respondent provided. However it was hoped that responses might be more frank and thoughtful on paper. In retrospect, some participants clearly spent more time with their responses than others who apparently completed their questionnaire quite hurriedly. 6. On the one hand, this convenience factor would suggest that participants in the study might be more predisposed to socializing and identifying with other Chinese. If this is so, then, as we will see later, the results may actually understate the degree of openness, which the second generation has to the broader society as students selected at random might have been even more open. On the other hand, locating the sample in this way may provide a more rigorous test of feelings of segmentation. 7. There is considerable variation in the way that first and second generation is distinguished in the literature. There is widespread agreement that foreign born who arrive as small children should not be considered first generation immigrants but the question is at what point do we mark the transition from childhood to adulthood (e.g. 12? 16? 18?). Zhou and Bankston (1998, p. 4), for example, call those arriving under 5 years of age as the second generation, those from 5 to 12 as the 1.5 generation, and those over 12 as the first generation. Park (1999, p. 163) calls all native born and the 1.5 generation the postimmigrant generation. The operational definition of second generation in this study includes all those who received their pre-post-secondary education in Canada and is similar to Portes and Rumbaut (2001, p. 23) who include foreign born children brought to the U.S. before adolescence as the second generation.

96

HARRY H. HILLER AND VERNA CHOW

8. Very few respondents used a hyphen in linking Canadian with Chinese regardless of the order in which the words appeared. It is unclear if this has any meaning. 9. A good illustration of the dilemma in understanding the meaning of the order in which this identity was expressed is contained in the following statement. “My first impulse would be to answer Chinese Canadian, but I always try to reanswer Canadian Chinese.” Future research needs to examine this identity issue much more closely. 10. Compare these results with an earlier study of university students by Frideres and Goldenberg (1977) prior to the wave of new immigration that found that most students specified their ethnic identity as “Canadian” and only 12% with a hyphenated identity. 11. Compare the classic position of the second generation that often felt embarrassed or alienated and rebelled against their heritage. 12. This is a form of ancestor worship where ripened fruit is left out in a dish for the benefit of the spirit of those who have died. 13. A few respondents answering affirmatively to marrying or dating a non-Chinese considered other Orientals non-Chinese and would restrict their choices to all Orientals. In short, they would choose beyond Chinese but would still try to restrict their choices to Asians suggesting that race was important to these respondents. 14. For example, Kelly (1998) speaks of the negative effects of blacks feeling highly visible and “under the gaze” of those in authority in a white dominant society monitoring their behavior. 15. The fact that the language’s existence is more and more dependent on its oral form is significant and does not portend a durable future. Compare Li (1998, p. 107) who also speaks of significant language loss in the second and subsequent generations in spite of a high percentage of persons in the Canadian population with Chinese as a mother tongue and language most often used at home. 16. The implication here is that beyond retaining fragments of the culture, much of Chinese culture may erode. Hoe (1976) and Friesen (1988) found a similar cultural acquiescence among Chinese youth. 17. Compare Lan (1993) who found that family ranked as the most important influence on self-identity. It is also important to note that a number of respondents made reference to wanting to preserve Chinese values or morals (particularly the role of family and respect for elders), the lack of which they considered a liability in the dominant society. 18. For example, dragon boat races or festivals have become major events in Canadian cities that often include non-Chinese as well. 19. Arguably Chinese voluntary associations could also be included in the private sphere. Fung (1998, pp. 118, 119) has shown how as Chinese Americans felt greater acceptance in American society, participation in such organizations became less important. 20. Breton et al. (1990, p. 259) come to a related conclusion in their study of Toronto that the Chinese do not emphasize ethnic retention and reconstruction and want to blend in at the same time that as a group they were weakly incorporated into Canadian culture. 21. There has been some discussion of the presence of a glass ceiling among Chinese in Canada but these young people are not yet in the labor market and may be unaware of such limitations. Also, compare these findings with “satellite children” where one or both parents are absent and there is much greater identity confusion (Tsang et al., 2003).

Ethnic Identity and Segmented Assimilation

97

REFERENCES Bacon, J. (1999). Constructing collective ethnic identities: The case of second generation ethnic Indians. Qualitative Sociology, 22, 141–160. Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology: An International Review, 46, 5–35. Boyd, M., & Grieco, E. M. (1998). Triumphant transitions: Socioeconomic achievements of the second generation in Canada. International Migration Review, 32, 853–876. Breton, R., Isajiw, W. W., Kalbach, W. E., & Reitz, J. G. (1990). Ethnic identity and equality: Varieties of experience in a Canadian city. Toronto: University Of Toronto Press. Chan, S. (1998). Race, ethnic culture, and gender in the construction of identities among second generation Chinese Americans, 1880s to 1930s. In: K. Scott Wong & S. Chan (Eds), Claiming America: Constructing Chinese American Identities during the Exclusion Era (pp. 127–164). Philadelphia: Temple University Press. Child, I. (1943). Italian or American? The second generation in conflict. New Haven: Yale University Press. Frideres, J., & Goldenberg, S. (1977). Hyphenated Canadians: A comparative analysis of ethnic, regional, and national identification of western Canadian university students. Journal of Ethnic Studies, 5, 91–100. Friesen, B. K. (1988). The maintenance of culture among Calgary’s Chinese. Multicultural Education Journal, 6, 4–20. Gans, H. J. (1979). Symbolic ethnicity: The future of ethnic groups and cultures in America. Ethnic and Racial Studies, 2, 1–20. Gans, H. J. (1992). Second generation decline: Scenarios for the economic and ethnic futures of the post-1965 American immigrants. Ethnic and Racial Studies, 15, 173–192. Gordon, M. M. (1964). Assimilation in American life. New York: Oxford University Press. Henry, F. (1994). The Caribbean diaspora in Toronto: Learning to live with racism. Toronto: University of Toronto Press. Henry, F., & Tator, C. (1999). State policy and practices as racialized discourse: Multiculturalism, the charter, and employment equity. In: P. S. Li (Ed.), Race And Ethnic Relations in Canada (pp. 88–115). Toronto: Oxford University Press. Henry, F., Tator, C., Mattis, W., & Rees, T. (1995). The colour of democracy: Racism in Canadian society. Toronto: Harcourt Brace. Hoe, S. B. (1976). Structural changes of two Chinese communities in Alberta, Canada. Ottawa: National Museum of Man Series. Hou, F., & Balakrishnan, T. R. (1996). The integration of visible minorities in contemporary Canadian society. Canadian Journal of Sociology, 21(3), 307–326. Hurh, W. M., & Kim, K. C. (1989). The success image of Asian Americans: Its validity and its practical and theoretical implications. Ethnic and Racial Studies, 12, 512–538. Isajiw, W. W., Sev’er, A., & Driedger, L. (1993). Ethnic identity and social mobility. A test of the drawback model. Canadian Journal of Sociology, 18, 177–198. Kalbach, W., Lanphier, M., Rhyne, D., & Richmond, A. H. (1970). Ethnogenerational factors in socioeconomic achievement in Toronto: The second generation during the 1970s. Toronto: Institute for Behavioral Research, York University. Kelly, J. (1998). Under the gaze: Learning to be black in white society. Halifax: Greenwood Press. Kibria, N. (1997). The construction of “Asian American”: Reflections on intermarriage and ethnic identity among second generation Chinese and Korean Americans. Ethnic and Racial Studies, 20, 523–544.

98

HARRY H. HILLER AND VERNA CHOW

Kibria, N. (2002). Becoming Asian American: Second generation Chinese and Korean American identities. Baltimore: Johns Hopkins University Press. Lai, D. C. (1988). Chinatowns: Towns within cities in Canada. Vancouver: University of British Columbia Press. Lan, K. S. (1993). The Chinese in Calgary. In: J. W. Friesen (Ed.), When Cultures Clash (pp. 167–188). Calgary: Detselig. Lee, S. (1998). Asian Americans: Diverse and growing. Population Bulletin, 53, 1–40. Li, P. S. (1998). The Chinese in Canada (2nd ed.). Toronto: Oxford University Press. Li, P. S. (1999). The multiculturalism debate. In: P. S. Li (Ed.), Race and Ethnic Relations in Canada (pp. 148–177). Toronto: Oxford. Lopez, D., & Espiritu, Y. (1990). Panethnicity in the United States: A theoretical framework. Ethnic and Racial Studies, 13, 198–224. Lyman, S. (1986). Chinatown and little Tokyo: Power, conflict and community among Chinese and Japanese immigrants in America. Associated Faculty Press. Lyman, S. M. (1977). The Chinese diaspora in America, 1850–1943. In: S. M. Lyman (Ed.), The Asian in North America (pp. 11–24). Santa Barbara: ABC-Clio Press. Lyman, S. M., & Douglass, W. M. (1977). Ethnicity: Strategies of collective and individual impression management. In: S. M. Lyman (Ed.), The Asian in North America (pp. 201–212). Santa Barbara: ABC-Clio Press. McCall, G. J., & Simmons, J. L. (1978). Identities and interactions. New York: Free Press. Nagel, J. (1994). Constructing ethnicity: Creating and recreating ethnic identity and culture. Social Problems, 41, 152–176. Neckerman, K. M., Carter, P., & Lee, J. (1999). Segmented assimilation and minority cultures of mobility. Ethnic and Racial Studies, 22, 945–965. Park, E. J. W. (1999). Friends or enemies? Generational politics in the Korean American community in Los Angeles. Qualitative Sociology, 22, 161–175. Park, R. E. (1950). The race relations cycle in Hawaii. In: Race and Culture. Glencoe: Free Press. Perez, L. (1994). The household structure of second generation children: An exploratory study of extended family arrangements. International Migration Review, 28, 736–747. Perlmann, J., & Waldinger, R. (1997). Second generation decline? Children of immigrants, past and present – A reconsideration. International Migration Review, 31, 893–922. Ponting, J. R., & Kieley, J. (1997). Disempowerment, justice, racism, and public opinion. In: J. R. Ponting (Ed.), First Nations in Canada: Perspectives on Opportunity, Empowerment, and Selfdetermination (pp. 152–192). Toronto: McGraw Hill Ryerson. Portes, A. (Ed.) (1996). The new second generation. New York: Russell Sage. Portes, A. (1997). Immigration theory for a new century: Some problems and opportunities. International Migration Review, 31, 799–825. Portes, A., & MacLeod, D. (1996). What shall I call myself? Hispanic identity formation in the second generation. Ethnic and Racial Studies, 19, 523–547. Portes, A., & Rumbaut, R. G. (2001). Legacies: The story of the immigrant second generation. Berkeley: University of California Press. Portes, A., & Schauffler, R. (1994). Language and the second generation: Bilingualism yesterday and today. International Migration Review, 28, 640–661. Portes, A., & Zhou, M. (1993). The new second generation: Segmented assimilation and its variants. Annals of the American Political and Social Sciences, 530, 74–96. Roberts, L. W., & Clifton, R. A. (1982). Explaining the ideology of Canadian multiculturalism. Canadian Public Policy, 8, 88–94.

Ethnic Identity and Segmented Assimilation

99

Rumbaut, R. G. (1994). The crucible within: Ethnic identity, self-esteem, and segmented assimilation among children of immigrants. International Migration Review, 28, 748–794. Rumbaut, R. G. (1997). Assimilation and its discontents: Between rhetoric and reality. International Migration Review, 31, 923–960. Statistics Canada (1998). The Daily, February 17. Stryker, S. (1985). Symbolic interaction and role theory. In: G. Lindsey & E. Aronson (Eds), The Handbook of Social Psychology (pp. 311–378). New York: Basic Books. Tonks, R. G., & Paranjpe, A. C. (1999). Am I Canadian, an ethnic, or an ethnic-Canadian?: Dilemmas of second generation immigrant youth. Working Paper #99–16, Research on immigration and integration in the Metropolis, Vancouver Centre of Excellence. Tsang, A. K. T., Irving, H., Alaggig, R., Chau, S. B. Y., & Benjamin, M. (2003). Negotiating ethnic identity in Canada: The case of the satellite children. Youth and Society, 34, 359–384. Tuan, M. (1999). Neither real Americans nor real Asians? Multigeneration Asian ethnics navigating the terrain of authenticity. Qualitative Sociology, 22, 105–125. Waters, M. C. (1990). Ethnic options: Choosing identities in America. Berkeley: University of California Press. Waters, M. C. (1994). Ethnic and racial identities of second generation black immigrants in New York City. International Migration Review, 28, 795–820. Yancey, W. L., Erickson, E. P., & Juliani, R. N. (1976). Emergent ethnicity: A review and reformulation. American Sociological Review, 41, 391–403. Zhou, M. (1997a). Growing up American: The challenge confronting immigrant children and the children of immigrants. Annual Review of Sociology, 23, 63–95. Zhou, M. (1997b). Segmented assimilation: Issues, controversies, and recent research on the new second generation. International Migration Review, 31, 975–1008. Zhou, M., & Bankston, C. (1994). Social capital and the adaptation of the second generation: The case of Vietnamese youth in New Orleans. International Migration Review, 28, 821–845. Zhou, M., & Bankston, C. (1998). Growing up American: How Vietnamese children adapt to life in the United States. New York: Russell Sage.

THE RADICALIZATION OF THE SELF – “BEYOND” GENERATIONAL ORDER: GERMAN CHILDREN’S LITERATURE AS A CASE STUDY Doris B¨uhler-Niederberger INTRODUCTION Childhood sociology as it has evolved from explicit critique of socialization sciences has developed two central concepts: “The child as (competent) actor” and the notion of “generational order.” It is above all the second concept that has not yet been fully dealt with within its sociological context. The term “generational order” is not just supposed to refer to ordered relations between (socially defined) age groups and their members, but also to a social order in general, as it is achieved by the ordered arrangement of age groups. From a historical perspective one can see that those efforts that aim at a disciplined society with small social control expenses do at the latest from the 19th century onwards concentrate on education and a well organized family and thus on a well ordered arrangement of age groups. It is an ordering process towards self-control, towards self-government as the most dense as well as discrete way of government. Until just some years ago such development appeared as an indispensable prerequisite of social order to those sociologists dealing with questions of childhood and growing up – at least as long as they assumed the perspective of socialization theory and sciences. Only the

Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 101–124 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10006-8

101

102

¨ DORIS BUHLER-NIEDERBERGER

absence or deficiency of such a generational order had any chance to become an important scientific question. This article will at first sum up how childhood sociology developed its two central concepts and thus opened the way for questioning socialization beyond anthropological assumptions for an analysis of processes of social ordering. Out of this I will focus on the question of the “social self.” As an authority that reflects, controls and presents itself the “self,” respectively the member of society that can be held responsible therefore, is a part of a certain (modern and western) attempt of social ordering and in many ways a variable quantity. This is how I will conceive it and this turn of the concept of the “social self,” which has at first been part of socialization theory, into a constructivist concept states the starting position of the empirical part. The ideals of self and self-government are subject to constant change, which is seen by some authors as the perfection of government, by others contrarily as growing egoism and erosion of society. I follow this change through children’s literature of the 20th century. One can see children’s literature in parts being a mirror of real circumstances, but first of all as a guidebook to the young reader containing guidance on working on the self, which children should do. The different solutions children’s books point out do sum up in their temporal order to form a trend in development: a radicalization of the self. The question may be asked in how far this development leads away from a generational order.

FROM CHILDHOOD SOCIOLOGY TO THE ANALYSIS OF SOCIAL ORDERING The sociology of childhood constituted itself through depositioning from the perspective of socialization. It criticized that the idea of becoming a social actor only by socialization and a kind of second birth is suppressing the competence of the child. In the beginning there were mainly scientists with the theoretical background of ethnomethodology as well as symbolic interactionism who accused the socialization approach with a paradox treatment of childish competence. On the one hand – thus was their understanding – the institutions and the arrangement of socialization demanded for children’s social competence, even took them for granted. These institutions required the explicit knowledge of rules and their validity. At the same time these rules were at no time clear and defined and thus in addition demanded the ability as well as the readiness of the children to find out the rules applying at that very moment (Davies, 1983; Mehan, 1974, 1979). It is especially about the rather subtle rules of interaction, who but produce the social actor’s categories of children as well as grown ups as diverse and

The Radicalization of the Self – “Beyond” Generational Order

103

complementary, as for example in the routines of greeting (Payne, 1976) or in the course of conversations intending different authorizations of access as well as ways of leading a conversation (Speier, 1976a). The studies showed, that children do in fact cope with these hidden institutional expectations and they also discovered, that even very small children are provided with fundamental social competences, for example the ability to meaningfully display social world or their self and to grasp its basic rules – which sums up altogether into taking part in the construction of social world (Denzin, 1971, 1977; Sacks, 1974). On the other hand – and this is exactly what was seen as the paradox – institutions, no matter if family or school, do ignore just these competences and this ignorance in actor’s competence with children would be valid for the perspective of socialization in quite a fundamental way (Jenks, 1982; Mackay, 1974; Speier, 1976b; Waksler, 1986). The sensitive focusing on minorities within the social sciences helped such theoretical fringe existences to gain popularity in the 1990s and influenced scientific approaches to childhood in sociology, as well as far beyond. In parallel with feminist criticism childhood-sociological criticism now accused earlier sociological theory and science not to have focused on children as actors (Ambert, 1986) or if they did so, to have always treated it as an actor with deficits, who was but to reach the status of an actor. Within a fundamentally possible reconstruction of the world only in perspective the voice and the experience of children was thus, just like that of women, locked out (Alanen, 1994), children were enclosed in an adult focused and patriarchal conception of the world (Leonard, 1990).1 Childhood is thus not discussed as an autonomous stage or situation in life but just as a time of preparation (Alanen, 1989, 1997; James & Prout, 1997; Qvortrup, 1993; Zeiher, 1996). The new childhood-sociological approach brought to life scientific studies that obviously set themselves apart from traditional contexts or questions of socialization. Some of these works shall be mentioned as examples: More than one project is focusing on the way children act competently in situations, which are educationally not structured as well as how they are able to use the conditions of urban space and time schedules (Behnken, 1990; Behnken et al., 1989; Mayall, 1994; Rabe-Kleeberg & Zeiher, 1984; Zeiher & Zeiher, 1994). Other studies concentrated on the cultural achievements of children and how they are able to use the supplies of the consumer society in their very own ways (Hengst, 1990, 2000), or how they create their own rules and interpretations of interaction amongst themselves (Corsaro, 1985; Corsaro & Eder, 1990). Again others did focus on socialization contexts, but discussed the achievements of children as competent actors, for example the negotiations children are having with their families (du Bois-Reymond, 1998), the responsibilities they also take over within (Alanen, 2000; Morrow, 1998; Solberg, 1990; Song, 1996; Zeiher, 2000) or the very own

104

¨ DORIS BUHLER-NIEDERBERGER

worlds they create at school or at kindergarten amongst themselves (Breidenstein & Kelle, 1996, 1998; Shiose, 1994, 1995; Strandell, 1997). All these studies present an important corrective of socialization sciences and create credibility and illustrative material for the starting premise of the competent actor, they prove his taking part in a social world, which has practically been the acting of children at all times. Thus childhood sociology has taken away the “natural,” the anthropological basis of the concept of socialization. The latter has generally assumed a “nature” of men, a fundamentally antisocial nature, which demands for an intense and systematic social finish. The child in deficit was to display this assumption. It stood for the human nature’s need for socialization and correspondingly its civilisation was supposed to take place in the state of life called “childhood,” which is especially intended therefore and therefore only (B¨uhler-Niederberger, 1998). But socialization is not a mere concept of sociology, it is social reality. It is an institutionalized reality of growing up and the deconstruction of pretended naturalness discloses the possibility of a sociological analysis of institutions and practices of socialization – without anthropological ballast and without a normative a priori. With its striking abstinence from topics of socialization the sociology of childhood still uses but a part of its analytical potential. Such a “sociology of socialization,” as it has now been made possible beyond its anthropological and normative take in, may ask fundamental questions on processes of social ordering and the establishment of government. It may therefore start with the term “generational order,” which has become quite common in childhood sociology in these days to grasp the fact of social and complementary construction of children’s and adult categories and the interaction between members of both categories. More than once it was postulated programmatically, that the focus must not be on the children alone, but on the generational relation, which has to be understood as a central element of society (Alanen, 2000; Honig, 1999; Qvortrup, 1987) and which has to be taken into view with its connection to gender and economic interests (Zeiher, 1996). The generational order establishes social inequality and places upon it socially accepted legitimacies: inequalities following age, and connected to this following gender and finally, via the supposed and socially defined quality of the upbringing following social status. But even if taking all these into consideration the conception of a generational order within a comprehensive social meaning, which shall be placed upon the concept, would have been analysed only in parts. Namely one can very generally see social order as generational. The arrangement of age groups does aim at nothing less but at a certain relation of individual and society. It especially does so when – and historically seen it has always done so since – it has become an arrangement of “socialization,” a systematic highlighting of

The Radicalization of the Self – “Beyond” Generational Order

105

differences between grown ups and children, a spatial and institutional separation of age groups in the interest of education, which tries to have an effect into the individual (Ari`es, 1992). The relation of individual and society that shall thus be achieved is of such kind, as it allows the individual to be addressed by social authorities whenever it concerns its own control, it is a self-governance. Modern educational childhood is not the only means of its establishment, but it is a crucial pillar. Incomprehensible without this request of a new control of the individual from within its closest vicinity namely by itself, why authorities and later on the state did put so much weight on the new arrangement of a long education focused on the child (B¨uhler-Niederberger & Tremp, 2001). The term “social self” has been stated ahistorically and normatively by socialization sciences. A certain self-authority has been stated as an indispensable “equipment” of the social actor. An ego, standing out from society, steady and united in a sense of identity has been seen as a goal of socialization and as a necessary prerequisite of social interaction and democratic relationships (Bilden, 1991; Krappmann, 1969). With the focus on generational order as self-control and thus as a mode of social control, the question on ideas of the self and on practices of its formation in educational institutions and contexts now becomes newly placed. Now a whole range of self-authority types may be thought of – each one constituting and being a cornerstone of a different relation of individual and society. Every change of conception and practices of self formation may be directly reconsidered in terms of its meaning for the social order. When suggesting a further development of childhood sociology by a partial, though revised grasp on the questions of socialization sciences – specifically the formation of the self – this does not happen (as it can be seen with other authors) to save the alliance of sociological childhood sciences and educational sciences, as it had become quite common in the 1970s and 1980s. It does also not happen while trying to maintain a subject term in the tradition of enlightenment or educational theory (see Zinnecker, 1996). Much more it is about connecting childhood sociology to that fundamental and central question of sociology for social order, the relation of individual and society.

THE SELF AS SOCIAL ADDRESS AND SOCIAL PROTAGONIST In the beginning of the 20th century theoreticians like Charles Horton Cooley, William James and George Herbert Mead introduced the concept of “the self” into social sciences. It is the perception of a human focusing on him. But this focus is no individual matter it is not a retreat from society. On the contrary it does always

106

¨ DORIS BUHLER-NIEDERBERGER

happen and is actually only possible in social contexts. The self is a “looking glass self” (Cooley, 1922) and the mirror, in which it questions itself is its opposites: concrete and significant opposites, which do mean a lot to the self, as well as abstract, much generalized opposites, a “generalized other” (Mead, 1934). The concept of the self means a significant achievement in terms of reflecting on society. Thus the dichotomy of individual and society becomes obsolete, in which thinkers of the 19th century expected to find the reasons and possible solutions for social problems and in which many sociologists and other social observers still see these reasons today.2 The self, viewing itself with social eyes, reflecting and thus positioning itself in social context as well as envisioning itself as something projecting and unmistakable, is individual and society at the same time. This should above all be taken into consideration when using the term individualization in the following. Individualization taken as an emphasis of individual difference is a mode of social affiliation, for those individuals embodying the social process called “individualization” are always and to a special extent “self.” This means they are reflexive humans who work out and present their uniqueness in a special way and thus it is not about the return to a pre-social being – which is controlled neither by social rules nor social affiliation – as well as about the return into a state of war “of all against all” as Hobbes calls it. In the course of the 20th century we have learned a great deal about the self as a social authority. Following Mead one would be tempted to see the concept of the self as a primarily anthropological insight into the constitution of man and the possibility of socialization. Scientists approaching this topic from a culturalist point of view have shown convincingly that actor authority in a society and the addressee of social demands are of changeable size. Thus David W. Sabean states about man in Lutheran times that he didn’t have a meaning of the self as a fixed centre of consciousness and that he didn’t keep any memories leading to the perception of personal unity. To him consciousness appeared to be a sequence of separate states reflecting the concrete social relationships of a person (2001, p. 148). From an ethnological point of view Clifford Geertz states the occidental vision of a person being “strictly outlined by a unique, more or less motivationally integrated and cognitive universe, by a dynamic centre of consciousness, of feeling, judgement and acting, organised as a separable whole standing out from other wholes as well as from a social and natural background.” For other cultures – for which he as well states significant differences from each other – he states this vision as being disconcerting (1987, p. 294). The “looking glass self” is thus not just socially driven by emerging in every single constituted biography through repeated research in front of the social mirror. It is a social invention, for the search itself as well as the fact that it is demanded and accounted for by society.

The Radicalization of the Self – “Beyond” Generational Order

107

Historians talk about the “discovery of the individual” (van D¨ulmen, 1997, 2001) or the “appearance of the individual” (Duby & Braunstein, 1990, p. 294) as a process which steps and elements they follow since the medieval ages. It was as well about aesthetical methods, like for example the cultivation of love, and aesthetical dealing with the inner person, as for example in collecting memories and the retreat into personal refuges. It was also about new forms of social interweaving, demanding consideration and restriction and therefore control, an observation and treatment of the person and its feelings, as well as cultivating its manners (Elias, 1976). But not just the intellectual striving for the self, for self-enjoyment and the expectance of social acknowledgment were the driving forces behind the social invention of a reflexive authority. Right from the beginning this development was also motivated by new modes of government. They addressed – and that was new about them – an individual that they built up (disciplinarily), for example by governing space and time in a way that snatched the individual’s behavior from the former dark, displaying it in the front of the controling gaze and urging to self correction (Foucault, 1977). Other governing instruments, taking a self for granted or constituting it during their application were for example the interrogation in front of a court or the painful interrogation and also confession. Inside of the subject they were in search for motives of action and demanded for an analysis of inner intentions. This they demanded of the actor as well as of social institutions; these could or should now take insight or remorse into account (van D¨ulmen, 1997, p. 39). Even God now claimed such reflexive authorities and charged conscience or sincere remorse to account. The history of its social invention displays clearly, that the self is by far an individual matter; it is not a space free from society. Much more it is placed at the powerhouse of social order. Modern government is bound to technologies of the self to a high degree. Self means “self-control,” in terms of governing others by social construction and the subsequent use of self-authorities as well as by governing of these authorities over themselves. Foucault has introduced the term of Gouvernementalit´e to show that subjects may control themselves through subjectivity and thus from the greatest proximity and the greatest efficiency possible (2000). But this is not just simple social stamping; it is also contrasting from it. The self is always a fundamentally social address, but at the same time it is also a social actor standing out, for reflecting oneself, producing and controlling oneself are always processes resulting in dissociation. The focus on the self can only be successful, if the own person is projected apart from the social background. Thus it inevitably leads “beyond class and social status” (Beck, 1994), beyond conventions as they apply to clearly defined social categories. This becomes apparent not just in the contemporary process of individualization, it has been going along with self-reflections since medieval times. Reflecting the

108

¨ DORIS BUHLER-NIEDERBERGER

own person and displaying and communicating this reflection, as for example in correspondence or literature, introduced questioning the system of estate (Schrader, 1996). Thus the self becomes a shaped as well as a shaping authority, a social addressee and actor, a reference to society as well as a simultaneous dissociation.

SELF IDEALS OF THE PRESENT – INDIVIDUALISM, DISPLAY, AESTHETICALIZATION At times sociologists diagnose – with widespread acceptance – on contemporary changes in social understanding and ideals concerning such authorities. The diagnosis of individualization may be the best known. It states a growth in meaning and emphasis of differences, of individual uniqueness, of the difference in nature of the self compared to all the others. This is interpreted as a possibility for liberation from conventions and rules of different groups and communities, to which one has been connected obligingly before – thus for an individuality apart from class and social level. The argument goes that this may become a compulsion, too, vis-`a-vis the multitude of options possible. This debate does always include the concern; it is about the erosion of social cohesion and about risky freedom (Beck, 1994). Following Richard Sennett (1977) it is not just the uniqueness and the total difference that is highlighted by that, but it is also authenticity, a correspondence between inside and outside that has to be achieved and displayed. Sennet states a growing importance of self display since the 19th century, which must now no longer be a display in terms of acting, but must be valid as authenticity. Herein Sennet sees the meaningful, in the compulsion to reveal the self and in the display of the inner self and thus in the loss of the art of pure acting. This goes along with a change or a decay of the public, which becomes a mere reflection of the self and a tyranny of intimacy. Finally an aestheticalization shall be diagnosed as contemporary development, a construction of the self via the taste, described initially as an invention of the 18th century by Flandrin (1982). Due to its demand taste leads to the innermost, the real truth of the individual and “good” taste is always connected to a certain social status: the elite has taste and its possession legitimates the affiliation to this group. In some cases, just as an acknowledgment of this association it enables a person to enter the elite.3 In this sense the proposal that aestheticalization in present times has again grown in importance does also convey the proposal that new processes of elite-building in connection with a treatment of self-ideals are underway (B¨uhler-Niederberger, 2002a, 2003).

The Radicalization of the Self – “Beyond” Generational Order

109

All these diagnoses regarding the change of self-images do imply the thesis that the display of individual uniqueness, i.e. the display of inner conditions and processes has grown in meaning. Goffman’s statement on “. . . the structure of the self becoming understandable seen under the aspect of display” (2000, p. 230), as well as his use of terms taken from the theatre – such as stage and backstage, props and interplay of the cast – becomes even more relevant, whereas it has to be stated that his “theatre” claims to be authentic and not mere role play.

SELF AND SOCIALIZATION – CHILDREN’S LITERATURE AS GUIDE AND MIRROR Self and socialization are connected inseparably. The generational arrangement of socialization with responsible adults and depending children as well as the emotional intensity which is supposed to connect the members of both categories (Donzelot, 1979; Sch¨utze, 1991) aims at the internalization of norms. It aims at their effects on the inner person. From the moment the self becomes a meaningful social quantity, it is always seen as a self in development, as a task and a permanent obligation to its formation. This at any rate is an element or even a key element of bourgeois qualities, as they emerge during the 19th century (Hettling & Hoffmann, 2000; Schlumbohm, 1998). It is connected to the thought that such a reflexive and controlling authority exists very much in deficit in individual development, i.e. in adolescence. Society wouldn’t be possible with such quantities and thus it founds the claim for necessary socialization, of educating social beings. Examining the shaping, educating work with the self of the child, children’s literature is helpful. Children’s literature tells young readers how children in literature find themselves, which wrong tracks and confusions may precede success. It claims to be at least in parts a display of reality and because childhood and education are normatively standardized social fields, fiction, underlying fewer compulsions of repression and whitewashing, may thus be more authentic than reality, as it could be reconstructed by oral history or interviews. But it also lays claim to be a guide by showing young readers media and methods to find their selves as well as desirable results and offering the possibility of identifying with their heroes. All of the books that are to be discussed in the following, show through their young heroes the treatment of the self, what they should and do carry out. They show the interests and demands of the social environment as well as the possibilities permitted and boundaries of asserting oneself. They provide an insight into reality and ideals in terms of the three main processes of the self that could be distinguished from each other at least theoretically: self-reflection, self-control, self-display.

110

¨ DORIS BUHLER-NIEDERBERGER

METHOD: SELECTION OF BOOKS AND CONTENT ANALYSIS Six books have been chosen for the analysis, all of which were published at different times in the 20th century in several editions and which were decorated with prices. Criteria for selection were their high degree of spreading in the German-speaking area and additionally the detailed nature, in which they described the processes of finding and forming the self. The books were taken from three different literary genres: detective stories, schoolboy stories and girls’ tales. This choice was made in order to gain a certain variety in the modes of self-formation displayed. All books were submitted to a qualitative content analysis. At first they were examined as pairs – according to their genre, looking for contrastive or similar elements in those two pieces written in different times. Then comparisons between the genres followed. To grasp contrasts or similarities, the heroes of each story were described as a “typus,” focusing on those features constituting them in order of their self-formation processes. The typological characterisation did initially start with the contrast: the young heroes were typified in their most marked difference. A type-formation like this is a procedure of classification in which – as the classification is defined – a number of features is assigned to a number of elements. The sum of elements is in most cases just one: the hero of the story (still some books do present more than one hero). The formation of types is a classificatory order having a relatively low demand for systematics. This means for the work in hand that the dimensions for describing the types were not always and strictly the same. Barton and Lazarsfeld (1955) have already suggested such a filing of data, to structure the qualitative analysis. It is a means of handling data that has been and is still used in qualitative studies, although this is not made explicit in most cases. In the analysis of data it works on the constant comparison of the relevant material, while the selection of data is a form of theoretical sampling in search for the greatest possible variety in the phenomenon at hand. The aim of treating data like that is to gain typological descriptions with substantial theoretical content. These three features: the constant comparison, the theoretical sampling and the search for theoretical concepts, which are grounded in the empirical data, are the basic features of qualitative research, understanding itself as a self-contained methodological approach (Strauss & Corbin, 1990). Barton and Lazarsfeld (1955) did also accept type-formations, in which discrete types had an unclear logical connection with each other, in which eventually every type is even described by a different bundle of variables. In this sense the classification carried out in my study is more complete: It characterizes the elements at least in parts by

The Radicalization of the Self – “Beyond” Generational Order

111

assigning the different categories of the same variables to them. The advantage of an “incomplete” classification through the formation of types such as this is: (a) in the conciseness of the types that can be gained: in the possibility to cope with the special features of a type without measuring it with dimensions which do not have much relevance for their special features. Thus (b) material can be taken into account that was not produced as comparable in the first place – and this does especially apply to the study at hand, for which material evolved as an artistic product. Still an incomplete classification b type-formation does point out a temporary state of research. The comparatively small sample is another reason for this preliminary nature: It is out of the question that the types of self-ideals and selfprocessing presented to the reader are by no means the full spectrum that has been presented in the course of the 20th century. Still the popularity of the books makes them become solutions as they were presented on prominent figures and brought close to the children. These figures, which will be presented to the reader of this article in the following, may it be the exemplary son Emil who is so much obliged to his mother (as are all of those similar child figures created by Erich K¨astner) or the boasting Bert, always focused on cheap success, were known to children of different generations. The books were given to the children as presents, they were exchanged among the young readers, they were made into movies, read at school etc. In this sense they are guides that have reached many children. Of course they do not represent the sole guide reaching the children. One has also to take into account the influence of family, of peers, of media and others. The types described here are not necessarily to initiate a further exploration of children’s literature, which would now cover a greater sample with the (rather problematic considering the artistic material) wish for higher systemacity. They should rather proof themselves useful in the study of real childhood and be a guide for an analysis of the multiple influences on the formation of the self and the normative demands that go along with it.

FROM PERSONAL OBLIGATION TO AN ABSTRACT HORIZON OF VALUES – TWO CHILDREN’S DETECTIVE STORIES Finding the self and self-development are not the main topic in these two stories about children as detectives. They are not beings full of deficits that have to be educated in the first place, for they act in a world of grown ups with a lot of competence. Still they do appear within the generational arrangement, i.e. as children who need the adults, as well as children in the process of development.

112

¨ DORIS BUHLER-NIEDERBERGER

Apart from the solving of the criminal case we get an insight into the task of self-formation, intended for them by the author. The first story, “Emil and the detectives” by Erich K¨astner and first published in 1928, is about a boy who is growing up alone with his mother. He is visiting his relatives in the big city of Berlin and is supposed to take some money to his grandmother, money the mother has saved up and that she owes to the grandmother as a means of support. Emil falls asleep on the train and is robbed. Together with a whole gang of children, organised exemplary, he is hunting the thief, finally catches him and gets back his money. In the second story, “Devilry – courier service rat’s tooth” by Andreas Schl¨uter, first published in 1991, Tina becomes the heroine. She, too, relies on her friends, a little group made up of two girls and a boy. Tina has just moved to Berlin, has met her new friends and together they start a courier service on inline-skates. During their work they come across a gang of Satanists. These people sacrifice animals and their leader is up to something even darker: He disposes off the dead animals in front of the doors of an animal experiment plant, from which he has been expelled before, and tries to impair its reputation. There is a professor just in this laboratory working on a method to make animal experiments superfluous. This is the place where the children first encounter one of the dead animals and they decide to clear up the case, which, after some detours and some scary experiences with occult practices they finally do. Both stories provide a sufficient number of moral points of orientation on which the detectives orientate themselves, and after which they organise their actions and judge themselves. Emil’s mother becomes the central authority in his moral world. He owes her gratitude and love and this is exactly how the figure is introduced right from the beginning. We first learn that he has spoiled his Sunday suit and that every time he does so he starts thinking about his hard working mother much too late, who provides him with food and the money to visit school. Here meagre conditions become a personal obligation. And this finally sets the criminal story underway: Emil is robbed of the money his mother has earned and thus has to get it back whatever happens. He shows that he is a real hero who will get it back. Emil also exercises a simple kind of self-control: whenever fear, tiredness or pain may lead him to act against his morals, he forces himself a little. As an image therefore one may see his way of pinching his calves, that prevents him from falling asleep and which he exercises on the way to Berlin, where he doesn’t want to fall asleep because of his mother’s money, or at school, where he wants to pay attention to get good marks – which he owes to his sacrificing mother. Tina in “Devilry” is just the opposite, being his professional colleague who is born 70 years later. She herself creates the horizon on which she orientates, permanently watching it and asking it and its stars, whose constellation and stance

The Radicalization of the Self – “Beyond” Generational Order

113

is changing due to this confrontation. The question of how she feels herself, which is not that easy to answer at all moments, plays an important role in this context. Looking for a starting point of the story we end up at Tina’s “F” in the last Math test. An “E” she thinks, would have been quite sufficient to express her failure, there was no need to give the worst mark possible. An “F” – written in big letters, red and with an exclamation mark, as if it was likely to overlook such a thing – is a severe humiliation, a hit below the belt. It is not likely that she will get into any trouble with her parents, who would rather comfort her with some encouraging remarks. It is the humiliation, as the book says, that she doesn’t know how to come to terms with. This is the problem that finally causes her idea to get a little bit of revenge from her teacher. The idea leads to the contact with a girl at her school who claims to be able to predict the content of the next Math test by means of a pendulum. She is the one who introduces her to the group of occult people who are responsible for the animal murders. Tina has some aids supporting her in the rather difficult work on her own self. Her room functions as a retreat and contains many books that she can consult for orientation. This is also where she writes her diary into which she pours out her heart for pages and pages. She has also got a rat, which she always wears on her body. Prohibited at school this is exactly the place where she needs her pet the most, the more she experiences failure and uncertainty. Trough this rat we get to know that Tina at times really wouldn’t know where to keep her hands if she wouldn’t be able to hold her rat. And this rat, as a living aid in self-orientation, is keeping the story alive. It falls into the hands of the occult group and adds some extra excitement to the story. Its name is Watson, telling us instantly that it is an important accessory for a young detective following in the footsteps of Sherlock Holmes. Whenever Emil orientates, evaluates and forms himself at personal obligation, it becomes quite clear what he has to do. Tina, whose work is self-referential for she gets many points of orientation out of herself, tends to be ambivalent, uncertain, her work a permanent search. Children like her match different adults. And thus we get to know about Emil’s mother that she is a widow who works hard for her son. She is in great worries about her son and overloads him with commands, suggestions and admonition. Although he is just going to Berlin for a visit of his relatives she cries on the way back from the station. She does not have her own life and so the personal addiction becomes mutual. To her Emil is the meaning of life and happiness. We get to know rather few details about Tina’s parents. They stick to the background and do not become main actors, they care about supplies, and they actually like to do that, but they also keep away from her whenever they think she wants to be left alone. They consider her unintelligible behaviour to be nothing but a first lovesickness and do not disturb her in her room. Nothing hints on personal addiction or mutual obligation. Supplying is a natural parents’ task

114

¨ DORIS BUHLER-NIEDERBERGER

and does not place any emotional obligation on the children. The parents gain happiness from their own lives. Thus their jobs have a different status, they do not mean everyday sacrifice. We get to know about the “superjob” Tina’s father just got at an architect’s office in Berlin and which is the reason the family had to move. Reasonable parents do not place unpleasant questions, as we get to know in the course of the story. This does not mean indifference of adults towards children, it could rather be termed a coupling of autonomous self-authorities, who want to and actually do cope with all problems by themselves, but who at the same time attentively scan if and when to intervene, following a certain trial and error principle.

FROM RASCAL WITH A HEART TO UNSCRUPULOUS SELF-PROJECTOR – TWO SCHOOLBOY STORIES “My name is Eugen” by Klaus Sch¨adelin was published for the first time in 1955 and experienced the 25th reprint in 2002. The story is about four 13-year-old boys from the city of Bern, three real rascals and a contrasting figure, a well-behaved and frightened “baby face.” Every chapter focuses on one childish prank which adds up to quite a collection: from stealing the aunt’s long, black theatre dress up to trying on a helmet in the city’s museum, which ends up in one boy being unable to remove it from his head and having to be carried out of the museum still wearing the helmet, as well as having to bring back this helmet in secret after finally having removed it. At closer look it is not just about breaking the rules or a lack of moral, which is to make the reader laugh. The pranks function as an acknowledgment of moral. Apart from all their cunning the little rascals carry their hearts in the right place and confirm the rules of morals. When in one prank they steal the false teeth of an unloved old teacher, they suffer from pity with the lonely old man and take the teeth back in secret. When stealing one of baby face’s numerous food parcels he is getting from home whilst the holiday camp, and sharing the food amongst all children, they are just teaching him a lesson in terms of egoism, the prank becoming an educational act. And this is exactly what the pranks are about over and over again: The do unmask false moral and respect the boundaries of righteousness. The rascal is acting self-education within the boundaries given, within the standards given and which he adapts for himself and where he separates the normative grain containing useless rules and false morals from the chaff of real humanity. This self-formation becomes reflected: by using the I-perspective of one of the rascals, which can not exactly be called introspection, but which is a kind of moral reasoning, in which he explains and justifies his actions. Again moral becomes above all a personal obligation: feelings of pity towards the old teacher, a

The Radicalization of the Self – “Beyond” Generational Order

115

bad conscience towards the aunt, a feeling of obligation towards the weaker “baby face.” One is obliged and pays back at times. Since 1987 – that is about thirty years later – many books about a rascal, Bert, appeared, written by the Swedish authors Anders Jacobsson and S¨oren Olsson. I refer to the volume “Berts collected catastrophes.” Bert is working on himself far more explicitly. He does not only appear as an I-narrator, but as a diary-writer, too. But does Bert develop a far more differentiated normative horizon than Eugen, similar to detective Tina, from whom we get to know again and again that she is brooding half nights and that she is questioning, is irritated and so on and that not just about the criminal case but about life and the world in general? No, Bert doesn’t have morals. He is occupied way too much with projecting himself on all levels: in the book, in his diary and in his life, about which he is telling us. It is always the presentation, its successful outcome or its failure, that is the criterion of evaluation and orientation of the own person. Therefore it is about sexes and ranking and both are connected to each other. Bert wants to be the successful boy, to be liked by the girls and well respected by the other boys. He hopes to be serious, adult-like and cool. That is the goal of his presentation, be it in life, in I-narration or in his diary. To keep a diary is a double-edged thing in this context: Bert decorates the cover with a skull to differentiate it from the girls’ diaries, and he keeps it secret by letting it slip away quickly, whenever the mother enters his room. Bert’s abilities in projecting, failing permanently, are displayed in a humorous and caricaturing way. The authors obviously think that children will happily be able to know a part of children’s culture, a problem they are confronted with in every day life. This can be seen as an authors’ criticism of how striving for acceptance may strictly govern self-formation, even if this is not stated in a moral way but just playing down the problem in a mere humorous way which could otherwise occupy the children. Seen like this the two schoolboys’ books are a kind of admonition. If the old admonishes to concentrate on the substantial norms, not to be well-behaved but to have the heart at the right spot, the young admonishes not to be misled by the appearance that is deceptive. The foil the schoolboy story sets itself apart from is the well-behaved philistine on the one hand, the well-styled successful-type on the other.

FROM CIVILIZATION TO AUTHENTICITY – THE GIRLS’ TALK Emmy von Rhoden first published “the Trotzkopf” in 1883. It is a classic in girls’ literature, reprinted and republished ever so often. It describes the education of an estate owner’s daughter, Ilse, from the age of fifteen to the time of her engagement.

116

¨ DORIS BUHLER-NIEDERBERGER

What happens here is the taming of the wild. The girl, described as impetuous and unrestrained has a good heart that is out of the question, but she has been treated with too much forbearance by her widowed father. Thus she has not learned to keep her stubbornness at bay, and she sticks to her wishes with all insistence and is not willing to make sacrifices. This makes such a striking weakness in her character – as we are told by one of her teachers – which finally conflicts with her marriage. This is exactly the reason why this teacher now has to work, because she used to have the same stubbornness in her youth. An important feature of stubborn character as shown here is excessive and indecent eating, uncontrolled body-movement and an overall indecent outward appearance. The soft but straight stepmother forces her to attend a boarding school. Here she learns how to eat in a decent way, to tame her appetite, to do fine needlework with patience, to keep her wild temper at bay, to take back oversized honesty in a courteous way. And she learns to account for herself in keeping the diary that her stepmother has given her. Success is at hand, in the end of the novel Ilse meets a young lawyer who asks her to become his “little wife”: What follows is Ilse’s engagement and the first kiss between the couple, to which she gives her consent after a last little touch of stubbornness. She has become “smaller,” the self-formation means civilisation, in the sense of Norbert Elias, keeping emotions at bay and rigorously suppressing wishes and drives. A hundred and twenty years later in Kristina Dunker’s novel, “Sleeping Beauty kisses,” Julia forms herself in a way that makes her connectable to love, too. She has to go through a different program: from “Julia Nobody” – as she calls herself in thought every time she realises her sister and father being not content with her, as she drifts away, no longer listening to conversations and not taking part and above all telling nothing about herself – she is to become “Julia Falke.” That means she is to become a girl who has characteristics that she can show her outside world, just as her well-accepted and very social sister, Ann-Kathrin Falke, does. Julia has to learn that it is not just “show” as she tends to think in the beginning, when she displays characteristic features, instead of taking refuge in her dreams. Display no longer has a negative connotation, as it had with rascal Bert, it is a prerequisite for the connection of self-authorities experiencing themselves as unique and thus claiming every other subject they get into contact with to be unique. Julia needs time and the encounter with a boy to learn all this. Before she is longing for the cat, her grandmother, for the beings that used to understand her when she said nothing, longing for the village world she had to leave behind when moving to Berlin with her sister and her father. Only when she learns to talk about herself and to make herself visible the love relation becomes possible. Julia’s display comes along in a very aesthetic way. She is writing a story for her friend, about a topic she has been thinking about for a long time, the escape of an alter ego from a cave into the daylight. She also learns how to pose as a model by her new boyfriend, who is a

The Radicalization of the Self – “Beyond” Generational Order

117

passionate photographer. Both points show in a very illustrative way the topic of displaying the own person, to make possible the social linking, the addressing in an individualized society.

CONCLUSION – FROM FAMILIAR OBLIGATION TO COUPLING OF SELF-REFERENTIAL SUBJECTS When looking in children’s literature for descriptions of processes and moments of the self, for self-reflection, self-control or self-projection we can find great changeability of this authority. For the decades gone it can be described by the words: from the good child, that brings joy and is the one and only happiness for its parents, to the child’s right to be an autonomous individual, which is not just its right, but also an obliging expectation. Parents as well as their children are taken up on their promise. They all have to care for the fact that the search for their own points of orientation is carried out with the intensity necessary and that signals about how matters stand and about success in their quest are being transmitted and received correctly. Thus it is – which shall be put ahead of all further comments – no laissez-faire and just in parts a plus in terms of freedom, it is self-socialization and this demands greater effort of all participants. The program of the “good child” used to be comparatively simple: The expectations and the happiness of the adults were moral points of orientation, after which action could be regulated, and a criterion of self-assessment. Though the simple program demanded rigorous discipline, as we can clearly see with “Trotzkopf” Ilse, but also with detective Emil, who owed the best grades, a clean Sunday suit and so on to his mother and still was completely in her dept. But it demanded little interpretive work: it was obvious what had to be done and what was good or bad. In the case of Eugen and his friends, i.e. in the book from the fifties, the self-control demanded is clearly reduced. Here, too, the personal obligation stays decisive in the end, but it turns out to be sufficient to have the heart in the right place. The good heart is placed above good manners, and thus authenticity is placed above civilisation. The time of paragons is over. Still full of pride of her successful detective son Emil’s mother can say: “No, actually he doesn’t have any shortcomings, my Emil” (p. 158) and simultaneously wipe her nose full of emotion. This will not be possible with the heroes to come. Weaknesses do inevitably belong to the self, the formation of which now stays an unfinished try. The program of the three contemporary children, rascal Bert, detective Tina and Julia Nobody is very demanding. By looking at themselves, by regulating and presenting their behaviour they only start looking for points of reference and standards for assessment that could be valid. They have to define them

118

¨ DORIS BUHLER-NIEDERBERGER

by themselves because exactly here the individual uniqueness starts to become real, the only person that can and must be introduced as such to others. Bert is taking it a little easier than the two girls when he limits himself to presentation and its effects, but just here one could see the criticism of the authors. His regular and funny failing in presenting is due to the fact that his presentation is not tailored to his own person, orientating at ranking instead of the possibilities of his own person. This is what the young reader is to learn. The new children correspond to new adults. The children don’t have to please; they don’t have to be the object of happiness. They don’t have to, because their parents have different pleasures and objects of happiness. The parents, too, are autonomous individuals. The new child is a component of the new family in which the members do not owe happiness to each other and everyone gets it for himself. This can be clearly seen in the example of Julia Nobody’s mother. She has moved away from the family and her daughter can only accept this, when she finally starts to “realise” herself, i.e. to define and display precisely. In her story about the escape from the cave, as Julia writes it, her mother is the first of the cave people to appear, the first who dares to escape and who tries to persuade the others. She has to pay for that with her life and is killed by the leader of the cave people. Still her escape is a manual for the escape of Julia Niemand’s alter ego, at first building it up in her fantasies and thus defining her dream world that cuts her off from everybody else. She finally presents it as an aesthetic self-product, as a written story inside of a story that finally enables her to get into contact with others. The autonomous individuals refer to each other. They do it permanently and it is always a try to be liable to each other, to get into contact. The investigational work is thus not just for the own self but also for others. By displaying they make themselves available to each other whenever possible. And still this combination fails at times. The detective’s parents, e.g. often tend to misinterpret the signals given by their daughter and the girl finally realises this even before her parents’ reaction, she gives further explanations, she corrects her presentation, sometimes deceptively, sometimes to make possible the correct connection. Even the compulsion for authenticity seems to be loosened, it is just important that a connection is made possible at all. So it is not just a new child, it is a new social cohesion, which is introduced here as a new reality and a new ideal. This represents a new development, from personal and culpable addiction in defined categories to loose coupling of autonomous individuals in no way representing anti-sociability but rather a new obligation towards ongoing trial. The transition can be described by the keywords “from domestic happiness to creative individual” (B¨uhler-Niederberger, 2002b). It means a gradual turning away from concepts of happiness, as they were promoted in earlier times, according to these where contentment in and around the domestic

The Radicalization of the Self – “Beyond” Generational Order

119

community were made a moral must and were a compulsion to gain happiness. This obligation stood above every right for individual features or taste, it matched the individuals to one another – being members of gender and age categories, who clearly structured behavior and mutual expectations. It was removed by the call to discover the personal self, personal likings and personal taste and thus as an individual created in that fashion being ready for social contact at any time. Just that is how children’s literature treats the topic. It makes it a constituent in the act of conversing society, displays it and brings it over to the young readers, not least giving detailed hints on its creation in personal cases. This new cohesion is not eligible just as one would like. It is a new structural pattern, a new connection of the individual with society, from which nobody can be spared. This becomes very obvious with Julia Niemand who is hooked up to a traditional world and thus wants to escape every new pattern. With all clarity she is encouraged again and again to be open to this new order. In how far is it also about an order beyond generational order? We do in fact see that the socialization of a radicalized-self places great demands upon the children and that grown-ups can only be of limited help. Thus the socialization of the self is in great parts self socialization. But this, too, is embedded into a generational arrangement. Still the adults keep the leading position, even if age roles are also treated individually, e.g. does parenthood no longer demand to live together with the family, and differences in competence between children and adults are emphasized less. Nevertheless should the occasion arise the adults are to push their youngsters towards demanding self-treatment and their own selftreatment lets them stride ahead. The parents of Julia and Tina do not just have the freedom of realising themselves due to their peculiarities, they are also obliged to do so. They have got jobs they like. The fathers of Tina and Julia have moved to the capital of Berlin in line with their professional career, as the reader gets to know. Julia’s mother does especially present herself as a pioneer on this quest for freedom, shying away from no risk. Still the adults stay true to their duties: with their experience in exploring and realising personal individuality, theirs and that of their children. The latter is shown in a study by de Singly (1996), in which parents see it as their task to discover the hidden talents of their children through offers and observation, to expand their success. The formation of the self is also kept up within a generational arrangement in the new “negotiation households” (du Bois-Reymond, B¨uchner & Kr¨uger, 1993) and “interactive pedagogy” (Fendler, 2001). Through its individualization the generational arrangement is not freed of demands of social control, it is not entirely negotiable and the generation that follows does not become the culturally dominant actor.4 In this new and subtler treatment of relationships between generations we still find a preparation towards social order, at least there is claim to do so. In this

¨ DORIS BUHLER-NIEDERBERGER

120

sense, too, the new kind of social cohesion stays a generational order. Does this imply more freedom or does it eventually even lead towards an erosion of society, because the self in its uniqueness is radically made the point of reference? One can clearly see that what is being additionally granted or even demanded in terms of uniqueness is simultaneously accompanied by the claim to display this uniqueness, so that it can be accounted for. The increase of freedom corresponds to an increase of investment into the coupling. When I said in the beginning, the self would be a reference to society as well as a deposition from it, one can now say that the radicalization of the self does in fact increase both aspects: deposition and reference to. Above all it is the new conception of happiness in which I can see an increase in freedom as well as an increase in government. Within a modern family everyone is taken up on his own promise in terms of happiness. He does not owe the others’ happiness, but much more his own. Happiness means above all successful selfrealization. You choose yourself how much of your self you realize, it is – as it is being suggested – up to you. But there is no excuse for not finding personal happiness or giving up the permanent search for it. The happiness of a skilful realization of the self becomes the engine and this engine has to go full blast. This is again a social order with increased efficiency of the individual, far more efficient than the disciplinary ages of the 17th and 18th century with their invention of supervision, described by Foucault (1977). Thus the radicalization of the self means more freedom and more government simultaneously – that is the paradox. This paradox may dissolve differently in individual experience in different places of society: it may be a gain of freedom or a pressure you cannot meet and from which you cannot distance yourself, because you are actually placing it upon yourself – that is the benefit and the danger of the radicalization of the self.

NOTES 1. See also Firestone (1973). ´ 2. As in the tradition of Emile Durkheim and Talcott Parsons, but also in contemporary pessimistic views of individualization, in which it appears as a loss of sociality. 3. See also the works of Bourdieu. 4. For a discussion of such ideas see Zinnecker (1999, p. 7).

REFERENCES Alanen, L. (1989). Von kleinen und von großen Menschen. Pl¨adoyer f¨ur eine Soziologie der Kindheit. Das Argument, 173, 79–89. Alanen, L. (1994). Zur Theorie der Kindheit. Sozialwissenschaftliche Literatur Rundschau, 17, 93–112.

The Radicalization of the Self – “Beyond” Generational Order

121

Alanen, L. (1997). Soziologie der Kindheit als Projekt: Perspektiven f¨ur die Forschung. Zeitschrift f¨ur Sozialisationsforschung und Erziehungssoziologie, 17, 162–177. Alanen, L. (2000). Childhood as a generational condition. Towards a relational theory of childhood. In: Researchers Attached to the Research Project “Child and Youth Culture” (Eds), Research in Childhood. Sociology, Culture and History. A Collection of Papers (pp. 11–30). Odense: University Press. Ambert, A.-M. (1986). Sociology of sociology. The place of children in North American sociology. Sociological Studies of Child Development, 1, 11–31. Ari`es, Ph. (1992). Geschichte der Kindheit. M¨unchen: dtv. Barton, A. H., & Lazarsfeld, P. F. (1955). Some functions of qualitative analysis in social research. Frankfurter Beitr¨age zur Soziologie, 1, 321–361. Beck, U. (1994). Jenseits von Stand und Klasse. In: U. Beck & E. Beck-Gernsheim (Eds), Riskante Freiheiten (pp. 43–60). Frankfurt: Suhrkamp. Behnken, I. (Ed.) (1990). Stadtgesellschaft und Kindheit im Prozess der Zivilisation. Opladen: Leske und Budrich. Behnken, I., du Bois-Reymond, M., & Zinnecker, J. (1900). Stadtgeschichte als Kindheitsgeschichte. Lebensr¨aume von Großstadtkindern in Deutschland und Holland um 1900. Opladen: Leske und Budrich. Bilden, H. (1991). Geschlechtsspezifische Sozialisation. In: K. Hurrelmann & D. Ulich (Eds), Neues Handbuch der Sozialisationsforschung (pp. 279–302). Weinheim, Basel: Beltz. Breidenstein, G., & Kelle, H. (1996). Jungen und M¨adchen in Gruppen: die interaktive Herstellung sozialer Unterschiede. In: K.-D. Lenzen & K.-J. Tillmann (Eds), Gleichheit und Differenz (Impuls Vol. 28, pp. 52–63). Bielefeld: Ver¨offentlichungen der Laborschule. Breidenstein, G., & Kelle, H. (1998). Geschlechteralltag in der Schulklasse: ethnographische Studien zur Gleichaltrigenkultur. Weinheim: Beltz. B¨uhler-Niederberger, D. (1998). The separative view. Is there any scientific approach to children. In: D. K. Behera (Ed.), Children and Childhood in our Contemporary Societies (pp. 51–66). Delhi: Kamla-Raj. B¨uhler-Niederberger, D. (2002a). Gl¨uck und Geheimnis – die Garderoben des gesellschaftlichen Akteurs. Essay. Sozialwissenschaftliche Literaturrundschau, 44(25), 71–86. B¨uhler-Niederberger, D. (2002b, 7–13 July). Times of the day – times of the year. Rituals and strategies of families. Paper presented at the XV World Congress of Sociology, Brisbane, Australia. B¨uhler-Niederberger, D. (2003). Nat¨urliche R¨aume oder Architektur der Kindheit. Die Verortung der Kinder zwischen generationalem Ordnen und Individualisierung. Neue Praxis, 33, 171–187. B¨uhler-Niederberger, D., & Tremp, P. (2001). Kinder und gesellschaftliche Ordnung – die generationale Grundlage moderner Demokratien. In: F. G¨uthoff & H. S¨unker (Eds), Handbuch Kinderrechte (pp. 37–67). M¨unster: Votum. Cooley, C. H. (1922). Human nature and the social order. New York: Charles Scribner’s Son. Corsaro, W. A. (1985). Friendship and peer culture in the early years. Norwood, NJ: Ablex. Corsaro, W. A., & Eder, D. (1990). Children’s peer culture. Annual Review of Sociology, 16, 197–220. Davies, B. (1983). The role pupils play in the social construction of classroom order. British Journal of Sociology of Education, 4, 55–69. Denzin, N. K. (1971). Childhood as a conversation of gestures. In: P. Hamilton (Ed.), George Herbert Mead: Critical Assessments (Vol. IV, pp. 51–66). London and New York: Routledge. Denzin, N. K. (1977). Childhood socialization. Studies in the development of language, social behavior, and identity. San Francisco: Josey-Bass.

122

¨ DORIS BUHLER-NIEDERBERGER

de Singly, F. (1996). Le soi, le couple et la famille. La famille, un lieu essentiel de reconnaissance et de valorisation de l’identit´e personelle. Paris: Nathan. Donzelot, J. (1979). Die Ordnung der Familie. Frankfurt: Suhrkamp. du Bois-Reymond, M. (1998). Der Verhandlungshaushalt im Modernisierungsprozess. In: P. B¨uchner, M. du Bois-Reymond, J. Ecarius, B. Fuhs & H.-H. Kr¨uger (Eds), Teenie-Welten. Aufwachsen in Drei Europ¨aischen Regionen (pp. 83–112). Opladen: Leske and Budrich. du Bois-Reymond, M., B¨uchner, P., & Kr¨uger, H.-H. (1993). Modern family as everyday negotiation. Continuities and discontinuities in parent-child relationships. Childhood, 1, 87–99. Duby, G., & Braunstein, Ph. (1990). Der Auftritt des Individuums. In: Ph. Ari`es & G. Duby (Eds), Geschichte des privaten Lebens. Bd. 2. Vom Feudalzeitalter zur Renaissance (pp. 471–589). Frankfurt: Fischer. D¨ulmen, R. v. (1997). Die Entdeckung des Individuums, 1500–1800. Frankfurt: Fischer. D¨ulmen, R. v. (2001). Entdeckung des Ich. Die Geschichte der Individualisierung vom Mittelalter bis zur Gegenwart. K¨oln: B¨ohlau. ¨ Elias, N. (1976). Uber den Prozess der Zivilisation. Bd. 1 und 2. Frankfurt/M.: Suhrkamp. Fendler, L. (2001). Educating flexible souls. In: K. Hultqvist & G. Dahlberg (Eds), Governing the Child in the New Millennium (pp. 118–142). New York, London: Falmer. Firestone, S. (1973). Nieder mit der Kindheit. Kursbuch, 34, 1–24. Flandrin, J. L. (1982). Pour une histoire du goˆut. L’Histoire aujourd’hui, 86. ¨ Foucault, M. (1977). Uberwachen und Strafen. Die Geburt des Gef¨angnisses. Frankfurt/M.: Suhrkamp. Foucault, M. (2000). Gouvernementalit¨at. In: T. Lemke, S. Krassmann & U. Br¨ockling (Eds), Gouvernementalit¨at der Gegenwart (pp. 41–67). Frankfurt a. M.: Suhrkamp. Geertz, C. (1987). Dichte Beschreibung. Beitr¨age zum Verstehen kultureller Systeme. Frankfurt: Suhrkamp. Goffman, E. (2000). Wir alle spielen Theater. Die Selbstdarstellung im Alltag. M¨unchen: Piper. Hengst, H. (1990). Szenenwechsel. Die Scripts der Medienindustrien in der Kinderkultur. In: M. Charlton & B. Bachmair (Eds), Medienkommunikation im Alltag (pp. 191–207). M¨unchen, New York, London, Paris: Internationales Zentralinstitut f¨ur das Jugend- und Bildungsfernsehen. Hengst, H. (2000). Vom Cow-boy zum Horse-girl. Working Paper 12, Child and Youth Culture. Odense: The Department of Contemporary Cultural Studies. Hettling, M., & Hoffmann, S.-L. (2000). Einleitung – Zur Historisierung b¨urgerlicher Werte. In: M. Hettling & S. L. Hoffmann (Eds), Der B¨urgerliche Wertehimmel. Innenansichten des 19. Jahrhunderts (pp. 7–22). G¨ottingen: Vandenhoeck and Ruprecht. Honig, M.-S. (1999). Entwurf zu einer Theorie der Kindheit. Frankfurt/M.: Suhrkamp. James, A. & Prout, A. (Eds) (1997). Constructing and reconstructing childhood. London, Washington: Falmer. Jenks, C. (1982). Constituting the child. In: C. Jenks (Ed.), The Sociology of Childhood (pp. 9–24). London: Batsford, Essential Readings. Krappmann, L. (1969). Soziologische Dimensionen der Identit¨at. Stuttgart: Klett. Leonard, D. (1990). Entwicklungstendenzen der Soziologie der Kindheit in Großbritannien. In: P. B¨uchner, H.-H. Kr¨uger & L. Chisholm (Eds), Kindheit und Jugend im internationalen Vergleich (pp. 37–52). Opladen: Leske and Budrich. Mackay, R. W. (1974). Conceptions of children and models of socialization. In: R. Turner (Ed.), Ethnomethodology. Selected Readings (pp. 180–193). Harmondsworth: Penguin Books. Mayall, B. (Ed.) (1994). Children’s childhoods: Observed and experienced. London: Falmer. Mead, G. H. (1934). Mind, self and society. Chicago: University Press.

The Radicalization of the Self – “Beyond” Generational Order

123

Mehan, H. (1974). Accomplishing classroom lessons. In: A. V. Cicourel et al. (Eds), Language Use and School Performance (pp. 76–142). New York, San Francisco, London: Academic Press. Mehan, H. (1979). Learning lessons. Social organization in the classroom. Cambridge, Massachusetts, London: Harvard University Press. Morrow, V. (1998). Understanding families: Children’s perspectives. London: National Children’s Bureau. Payne, G. C. (1976). Making a lesson happen. An ethnomethodological analysis. In: M. Hammersley & P. Woods (Eds), The Process of Schooling (pp. 33–40). London: Routledge and Kegan. Qvortrup, J. (1987). Introduction. International Journal of Sociology, 17 (Special Issue: The Sociology of Childhood), 3, 3–37. Qvortrup, J. (1993). Die soziale Definition von Kindheit. In: M. Markefka & B. Nauck (Eds), Handbuch der Kindheitsforschung (pp. 109–124). Neuwied: Luchterhand. ¨ Rabe-Kleeberg, U., & Zeiher, H. (1984). Kindheit und Zeit. Uber das Eindringen moderner Zeitorganisation in die Lebensbedingungen von Kindern. Zeitschrift f¨ur Sozialisationsforschung und Erziehungssoziologie, 4, 29–43. Sabean, D. W. (2001). Selbsterkundung, Beichte und Abendmahl. In: R. v. D¨ulmen (Ed.), Entdeckung des Ich. Die Geschichte der Individualisierung vom Mittelalter bis zur Gegenwart (pp. 145–162). K¨oln: B¨ohlau. Sacks, H. (1974). On the analysability of stories by children. In: R. Turner (Ed.), Ethnomethodology. Selected Readings (pp. 216–232). Harmondsworth: Penguin Books. Schlumbohm, J. (1998). Constructing individuality: Childhood memories in late eighteenth-century. Empirical psychology and autobiograpy. German History, 16, 29–42. Schrader, F. E. (1996). Die Formierung der b¨urgerlichen Gesellschaft, 1550–1850. Frankfurt a. M.: Fischer. Sch¨utze, Y. (1991). Die gute Mutter. Bielefeld: Kleine. Sennett, R. (1977). The fall of public man. New York: Knopf. Shiose, Y. (1994). Nous et les autres dans une classe au Qu´eb´ec. Des univers parall`eles. Anthropologie et Soci´et´es, 18(1), 77–92. Shiose, Y. (1995). Les loups sont-ils Qu´eb´ecois? Collection “Mutations et Soci´et´es”. Laval: Presses de l’Universit´e. Solberg, A. (1990). Negotiating childhood. Changing constructions of age for Norwegian children. In: A. James & A. Prout (Eds), Constructing and Reconstructing Childhood (pp. 118–137). Basingstoke: Falmer. Song, M. (1996). “Helping out”: Children’s labour participation in Chinese take-away businesses in Britain. In: J. Brannen & M. O’Brien (Eds), Children in Families. Research and Policy (pp. 101–113). London, Washington, DC: Falmer. Speier, M. (1976a). The Child as a conversationalist. In: M. Hammersley & P. Woods (Eds), The Process of Schooling (pp. 98–102). London: Routledge & Kegan. Speier, M. (1976b). The adult ideological viewpoints in studies of childhood. In: A. Skolnick (Ed.), Rethinking Childhood. Perspectives on Development and Society (pp. 168–186). Boston: Littler, Brown. Strandell, H. (1997). What are children doing? Paper presented at the International Conference “Childhood and Children’s Culture”. Esbjerg, Denmark (30 May–2 June 1997). Strauss, A., & Corbin, J. (1990). Basics of qualitative method. Newbury Park, CA: Sage. Waksler, F. C. (1986). Studying children. Phenomenological Insights. Human Studies, 9, 71–82. Zeiher, H. (1996). Kinder in der Gesellschaft und Kindheit in der Soziologie. Zeitschrift f¨ur Sozialisationsforschung und Erziehungssoziologie, 16, 26–46.

124

¨ DORIS BUHLER-NIEDERBERGER

Zeiher, H. (2000). Hausarbeit. Zur Integration der Kinder in die h¨ausliche Arbeitsteilung. In: H. Hengst & H. Zeiher (Eds), Die Arbeit der Kinder. Kindheitskonzept und Arbeitsteilung zwischen den Generationen (pp. 45–70). M¨unchen: Juventa. Zeiher, H., & Zeiher, H. (1994). Orte und Zeiten im Leben der Kinder. M¨unchen: Juventa. ¨ Zinnecker, J. (1996). Soziologie der Kindheit oder Sozialisation des Kindes? Uberlegungen zu einem aktuellem Paradigmenstreit. In: M. S. Honig, R. Leu & U. Nissen (Eds), Kinder und Kindheit. Soziokulturelle Muster – Sozialisationstheoretische Perspektiven (pp. 31–54). Weinheim: Beltz. Zinnecker, J. (1999). Children as agents of change: conceptualising the process of (re-) producing culture and society between generations. Sociological Analysis, 2, 1–14. Children’s Literature Sch¨adelin, K. (2002/1955). Mein Name ist Eugen. Z¨urich: Theolog. Verlag (25. Aufl.).

CHILDREN’S VIEWS ON CHILDREN’S RIGHTS. A COMPARATIVE STUDY OF SPAIN AND ITALY Angelo Saporiti, Ferran Casas, Daniela Grignoli, Antonio Mancini, Fabio Ferrucci, Marina Rago, Carles Alsinet, Cristina Figuer, M`onica Gonz´alez, Mireia Gus´o, Carles Rostan and Marta Sadurn´ı INTRODUCTION In the beginning the Philosophers were the ones to school us on how to consider and treat our children. In Plato’s Republic, Socrates sketched a place where the nursing and parenting of children is carried out communally (Platone, 1994, p. 172). Not much later, instead, Aristotle claimed that children belong to their own parents inasmuch as they beget them (Aristotele, 1999, pp. 345, 18–24). Much later, first Augustine, then Thomas, and later still Locke and Hobbes, Rousseau and Kant, all had a great deal to say on children and parents, as well as on children’s status (Archard, 1993, Chap. 1; Blustein, 1982). The last great master who told us who children are and how they should be treated is perhaps John Stuart Mill: children are immature beings who cannot have the very same rights and liberties as adults (Mill, 1910, p. 73). Nowadays we have no great maitres a` penser to lean upon anymore, and even if we still turn back to our ancient guides, we mostly rely upon the power of Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 125–152 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10007-X

125

126

ANGELO SAPORITI ET AL.

law. It is striking how much has been legislated on children’s social status and rights, both on the national and the international levels.1 However, it is even more striking how much physical and moral suffering still surrounds childhood, and how many hesitations – if not nuisances – still raise from any speaking about the prospect that children should have the same rights as adults, not only the right to be nurtured, protected and parented – which is obvious. It seems as if neither the enlightened thought of the masters of the Western society, nor the power of law have been enough to guarantee children those rights we are so proud of as among our highest achievements. On the other hand, it may seem that our need to resort to law in children’s matters is the implicit avowal of our crushing failure to recognize childhood as a constituent part of society (Raes, 1997). As a matter of fact, things are better now than ever before. In the last century something changed. Indeed, the negative feelings previously expressed arise from the slow but constant awakening to a new sentiment of childhood, in individual and collective conscience. The need to revise our conception of childhood and children is underway, as well as the necessity to change, if not eliminate once and for all, that subordinate – someone would say “oppressive” – relationship which always has characterized the indissoluble link between adults’ and children’s worlds. Such a process has developed along two different lines, of which just one is under discussion as regards its claim to underpin a new social status for children and childhood. On the one hand, never before have children been assured the basic needs to grow up and develop all their physical, moral and intellectual potentialities. Nowadays no one would deny the right of the child to be protected, cared for, and supported. Hence the growing centrality of childhood in contemporary societies and, accordingly, the undoubted development of children-centered family policies, as well as various forms of children tutelage, protection and promotion. On the other hand, all this does not seem to be enough. The still lasting shameful (for us) and wretched (for the children) conditions of many children in the world (including children in the so-called advanced societies), highlights the necessity to consider rights. Surely, making reference to rights could just be an exploitable move to better children’s conditions through normative implements; the rights discourse can also acquire the meaning of an ideological radical demand of equality with respect to adults, independent of any kind of strategic or practical calculus.2 Otherwise, in a more sophisticated reading, it could also be a purpose-built “rhetoric” (O’Neill, 1996, p. 36). Be that as it may, what matters is that right now the language of law and rights has become a typical way to deal with children’s matters, and to almost dominate the social and moral discourse on children all over the world, not only in Western societies. It is not by chance that any recent discussion about children has

Children’s Views on Children’s Rights

127

as its main point of reference the 1989 United Nation Convention on the Rights of the Child (UNCRC). But it is exactly the resort to this language that has opened again the issue about the “nature” and the status of the child raising a profound debate involving scholars in different fields. Naturally, because “taking children’s rights seriously”3 means opening a Pandora’s box and getting oneself into a fix of disciplinary and interdisciplinary questions, stretching the subject well beyond its starting point. All this – and this is the most explosive element, as it is the most controversial – is focused on the child, i.e. on the possibility/impossibility that a subject, who has been considered for such a long time (but still now) as immature, in need to be cared for, dependent on adults, incapable of understanding and promoting his/her best interest . . . can be a rights-holder capable of exercising his/her rights.4 In addition, to make the issue even more complex and intricate, there is the growing upheld and championed demand for children to take personally active part in social life and in the promotion, diffusion and tutelage of their own rights. In order to achieve this, at least a condition needs to be satisfied – children should be able to perceive and be aware of their own rights. The supporters of this position implicitly assume that only if such condition is met, children could empower themselves and exercise self-advocacy, thus moving from being passive recipients of adult care and control to active participants in the Great Game of society (Clark, 1989). This paper examines children’s perceptions or views of their own rights, taking for granted a general assumption and a presupposition. We assume that the full extent of human dignity includes both the capacity to be aware of and to express in different ways our rights, be them moral, legal or any other kind of rights; in addition, we presuppose that, in some way or another, children do have both capacities.

CHILDREN’S RIGHTS: INWARD PERSPECTIVES5 Until the beginning of the 1980s, a social representation of children as not-yet “persons” was still paramount among social sciences. Children were not considered full-fledged individuals as the adults were. Up to the coming-of-age, the child was deemed an immature being, incompetent, unable to make proper judgments about his/her own interests, incapable of taking care of him/herself, of expressing his/her wills and wishes, and so forth. Children and childhood were under an utter guardianship by the adults’ world, both from a conceptual point of view and in societal practices. Briefly stated, they had any role in playing the Great Game of society but to “wait and learn,” wait and learn to become adults.6 It therefore was truly unthinkable of speaking of “children’s rights.” Rights and their correlate

128

ANGELO SAPORITI ET AL.

language were exclusively kept for adults. Since children did not have rights, it was even less conceivable for them to express any kind of thought or deliberation about their rights. The way things stand, no one could oppose that the UNCRC has had the effect of a rocket propeller for promotion, advocacy and research on childhood, especially as far as the rights issue is concerned (Small & Limber, 2002, pp. 65–66). This does not mean, however, that formerly research on children’s rights and children’s views of their rights was totally missing. Indeed, well-known scholars studied children’s point of views on their rights even earlier than 1989, especially in the associated fields of legal concepts and witness before the Court.7 In all likelihood, this was a first step toward a big change: researchers started to doubt the validity of the “standard model” of the social representation of childhood. Besides, Piaget (1932) and Kohlberg (1981)’s work on the cognitive development of children had already paved the way for an enquiry into children’s moral thinking, a topic which is close-knit to justice and rights.8 Another step toward a changing image of childhood came from social dynamics that paralleled changes in the academic field. Slowly but steadily, a new sensibility gave way to a progressive new social representation of childhood according to which children are not just an “object” of attention and protection but also “subjects” entitled to and holding rights. However, with regard to our concern in this paper it is possible to say that research on children’s views of their rights started at the end of the 1970s. Melton (1980, 1983), for example, investigated the development of reasoning about children’s rights of children aged 4–13. Among the findings of Melton’s first study, it is not only age but also the socio-economic status of the family that clearly influences such development. School, on the other hand, seems to be less relevant than the cognitive development and social status. The study by Melton and Limber (1992) reinforced the conclusion that if age is the most important determinant of children’s conception of rights, social class is a very strong predictor of children’s reasoning about rights. Later, Cherney and Perry (1994) used Melton’s questionnaire with a small-size sample of children from Geneva (Swiss) and Nebraska (U.S.) to check if cultural differences are more relevant than social and economic factors. The results show that Melton’s hypothesis on the relevance of economic status is the best predictor of children’s views on their rights, even though research using wider samples is in need to assess if cultural differences count more than Melton and Limber believe. However, during the last decade a certain consensus came out: children and youth reasoning about rights seems to be strongly influenced by the social context. Furthermore, the idea that the children’s way of thinking about rights reflects the level of comprehension of the rights that are actually exerted in their life gained points (Ocha´ıta et al., 1994).

Children’s Views on Children’s Rights

129

Various scholars have underlined the different progression in the development of the reasoning about rights to provision and protection with respect to those concerning self-determination. However, some (e.g. Ruck et al., 1998) underline that due to a lack in empirical research, we do not yet understand those progressions. Regards this matter, Melton and Limber (1992) convey a different position. Comparing samples from three American States and from Norway, they noted that children follow a general progression with precise paces but different content, in the sense that while for American children self-determination is more relevant than provision and protection, for the Norwegian it is the contrary. Such results cast doubts about the hypothesis that the development of children’s conceptions of their own rights follows universal stage or steps. The starting of the debate on children’s views of their rights opened at least two new main lines of investigation. The first aimed to deepen the knowledge of the mechanisms and processes underlying the understanding of rights at different ages (Sparks, Girling & Smith, 2000). The second, places much more emphasis on the methodological aspects of the data gathering on children, at the same time putting forth new epistemological and ethical questions. This line of thought does not limit itself to children’s rights, but extends its criticism to a general reformulation of rationale research on children in social sciences (Backe-Hansen, 2003; Walker, 2001). The master challenge is to understand the children’s perspective, not only for the sake of knowledge but for practical and social ends as well. The main ends to be pursued are in fact both the prevention of the youngsters’ social exclusion and, above all, their consideration in all the social dynamics concerning their social participation (Casas, 1997a). Indeed, according to Melton and Limber (1992), the study of children’s rights from their own perspective goes beyond scientific interests. Furthermore, the very same authors stress, at the same time, the fact that the personal and direct involvement of children constitutes a key element in the implementation of the UNCRC. We have already defended the importance of such efforts for the promotion of democratic values (Casas, 1997b); in addition, we can underline the positive and beneficial psychosocial effects for all those involved in such participation (Chawla, 2001; Tapp & Levine, 1974; Walker, 2001). Even though research on adults’ and children’s views on various children’s rights matters has already produced a certain amount of knowledge, we agree with Small and Limber (2002, p. 70) when they say that “Social scientists should continue to pursue research to assess children’s perceptions of their rights at different ages and under varied circumstances,” and that research “also is needed to illuminate children’s subjective understanding of various constructs that are basic to children’s rights.”9

130

ANGELO SAPORITI ET AL.

These remarks hold especially as far as research on children, their own parents and their own schoolteachers is concerned. Indeed, research addressing from a global point of view the whole context in which children are requested to speak about their rights, is still largely demanded.10 As a matter of fact, the study we are going to present is a small contribution along this line. We start, too, from the premise that it is necessary to continue to study and deepen the different perspectives of adults and children with regards to children’s rights. But we think that the best results can be reached enquiring simultaneously the three “subjects” involved in the children’s rights “affair” – the very children, their parents and their school-teachers – and bearing in mind that we must give the same relevance both to children and adults, without blaming children when their opinions do not fit adults’ and researchers’ wishes, expectations or models.

CHILDREN’S PERCEPTION OF THEIR RIGHTS IN ITALY AND SPAIN Our main objective is to enlarge the existing knowledge on children’s perceptions of their rights, as well as those of their parents and schoolteachers. Following the previous considerations, we started from the general hypothesis that the respect of human rights is a phenomenon linked to the social and cultural milieu and, at the very same time, to the experiential life during the cognitive development of any human being. The exercise of certain rights, and the way we do it, have both a remarkable influence as to thinking and understanding the world of rights. We therefore decided to explore the perception that children have of their own rights, inquiring into their most significant living milieu – i.e. family and school.

The Research Setting The research has been carried out in Molise (Italy) and Catalonia (Spain). (See map). The two regions were chosen because they were considered to be similar from a socio-demographic point of view. Molise is the second smallest of the Italian administrative and political districts, located in the South-Center of the peninsula between the Apennines and the Adriatic Sea. Molise has only 300,000 inhabitants, and children aged 0–14 comprise 15.2% of the total population. Molise is divided into two provinces (Campobasso and Isernia) and the population is scattered into two-thirds in villages with less than 10,000 inhabitants, while only three towns have from 20,000 to 50,000 inhabitants (Campobasso, Termoli and Isernia). As to the economic

Children’s Views on Children’s Rights

131

Map 1.

structure, 15.6% of the population is active in agriculture, 27.8% in industry, and 56.6% in services (Map 1). The Spanish autonomous region of Catalonia is much bigger than Molise but the research has been carried out in the two provinces of Girona and Lleida (respectively 560,000 and 362,000 at 2001). According to the 2001 census, 14.5% and 13.3% of the population, respectively in the province of Girona and the province of Lleida, are younger than 15. Both provinces are more industrialized and urbanized than Molise; both, however, are also big markets for agricultural products. As to the 2002 economic structure, in the province of Girona 4.3% of the population is active in agriculture, 35.4 in industry, and 60.3 in services, while in the province of Lleida the data are, respectively, 13.0, 31.5, and 55.6%. Research Design and Instruments Following the antecedent considerations, in order to explore and to assess children’s perception of their rights at different ages and under varied circumstances, we

132

ANGELO SAPORITI ET AL.

envisaged a research strategy involving three different sources of data: children of both sexes aged 9–13, their parents, and their school-teachers. Data have been gathered through three different questionnaires, each one addressed to one of the three previously mentioned categories of subjects. Albeit similar in their substantive content and scope, the questionnaires are, of course, different. The children’s questionnaire. Apart from the ordinary background information (sex, age, school . . .), the children’s questionnaire includes a set of sixteen everyday, concrete situations implying moral dilemmas, with a possible solution proposed by the researchers. (See below). Children were expected to say whether they agree or not (“yes/no” answers) with the proposed solution. In eight out of the sixteen situations the child was asked to write down in open-ended format the reason for the given answer. The questionnaire includes three more “yes/no” dichotomized items (Do you know what a right is? Do children have rights? Do you know which the rights of the children are?) and two additional open-ended items asking children to say who has rights and what children’s rights are. The parents’ and teachers’ questionnaires. The parents’ and teachers’ questionnaires have many similar items, but only a few are strictly equivalent to those in the children’s questionnaire. Nonetheless, both include items that are very close in their scope to the ones presented to the children, in order to explore what kind of relationships, if any, there are between children’s, parents’ and teachers’ views of children’s rights. The majority of the items of both questionnaires are graduated according to a five-point scale (ranging from 1 corresponding to the lowest intensity of parent’s and teacher’s answer to 5 corresponding to the highest). The construction of the children’s questionnaire took quite a long time. In the beginning the Catalonia team drafted a first version of it based on previous pilot testing with a long list of items related to the rights recognized to children in the UNCRC. Efforts were invested in having a list of dilemmas referred to the so-called UNCRC 3 “Ps” (Provision, Protection and Participation rights), as well as to the right to self-determination. The questions have been posed both in positive and negative terms: dilemmas based on clear negative situations, potentially damaging somebody, and dilemmas based on non-negative situations, i.e. on situations in which somebody’s rights could be promoted. In order to have as many items as possible, the first step was based on personal creativity of team members and on long group discussions about pros and cons for each specific formulation. The subsequent step was testing children of different ages, and then team group discussion again, aiming at improving language and the whole formulation of the dilemmas. All this was done in order to let children understand sentences in a clear and unique way, making the situations as much concrete as possible. Finally, the Catalonia team selected the “best set” of 16 dilemmas, considering all aspects.

Children’s Views on Children’s Rights

133

Due to obvious reasons, both research groups agreed to have a short questionnaire. We therefore decided to include open-ended items only for half of the dilemmas. Once the Catalonia team had a final proposal, it was sent to the Molise team and some additional testing was made with children from Molise. After that, a two-day meeting was organized both in Girona and in Molise with the members of the two research groups and all testing results were discussed. After some changes and improvements, the final version of the questionnaire was translated into Italian and Catalan languages and a final pilot testing was conducted. Since the two languages have common Latin roots, semantic adaptations were minor, and it was very easy to agree on the final version of the three questionnaires and their equivalence.11

The Sample In each region we selected a number of towns and villages with overall similar socio-demographic characteristics and then we produced a list of all schools with children aged 9 to 13. The Molise sample included 93 nine-year-old children, because they were regular students of the same primary school level being clearly homogeneous with the other students of their groups. Next we selected those schools whose student population could be considered the most representative of the characteristics of the majority of the families in the town, excluding a few schools located at both extremes of the continuum: the elite schools and those from the poorest communities. Therefore, we had two samples in which the majority of children came from middle class families, according to the standards of each country (lower, middle and upper middle class). From the final list of schools we randomly selected a set of them in which we could hand in a number of questionnaires for each age group. We then reported our purposes both to the Principal and parents’ associations, and we proceeded in accordance with regular ethical guidelines, providing questionnaires to children in each country. We randomly selected a number of classrooms in order to have a sample of all age groups from the selected schools, and we asked for the cooperation of the responsible teacher/s of each classroom. After that, the children were asked for collaboration and were informed that the data would be treated confidentially and that they were free to refuse participation. The questionnaires were administered to the children in their regular classroom. One or some of their usual teachers and one or two researchers were present during the administration and clarified questions posed by children. The session usually lasted about 45–50 minutes for

134

ANGELO SAPORITI ET AL.

the youngest and about 30–35 minutes for the oldest. At the end of the session we gave each child a letter and a questionnaire for their parents in a sealed envelope to be delivered to them by hand. They were asked to return it to the teacher within one week, again in a sealed envelope. Parents were asked to cooperate in the research and they were told that the questionnaire could be answered by one, or by both of them together. The parents’ questionnaires had a code, so that they could be paired with each child’s questionnaire. In their questionnaire, parents were requested to answer in reference only to the child who had answered the school questionnaire whose name was marked on the form. Teachers from the selected schools were also asked to answer a specific questionnaire, which had many equivalent items with the parents’ questionnaire, in order to compare their answers in relation to the same child population. Table 1 displays the final sample, with its main background variables. From Table 1 we can see that while the two children samples are rather akin as to sex and residential area, the age distribution is a bit different, especially as far as the two extremes are concerned. The Catalan sample has no children aged 9, while the Molisan has fewer thirteen-year-olds.12 Table 1. The Structure of the Samples. Catalonia (Ca) # Children Boys Girls 9 10 11 12 13

Molise (Mo) %

#

%

645 315 330 – 115 218 220 92

100.0 48.8 51.2 – 17.8 33.8 34.1 14.3

969 493 476 93 282 366 208 16

100.0 50.9 49.1 9.6 29.2 37.9 21.6 1.7

Urban areas Rural areas

434a 211b

67.3 32.7

641c 328d

66.1 33.9

Parents School teachers

400 66

62.0e

844 192

87.1e

a Girona

and Lleida (70,000 and 112,000 inh., respectively). communes (4 having have less than 5,000 inh.). c Campobasso (50,000 inh.), Isernia (20,000 inh.), Termoli (30,000 inh.). d 8 communes with less than 5,000 inh. e % on children questionnaires. b5

Children’s Views on Children’s Rights

135

CHILDREN’S VIEWS OF THEIR OWN RIGHTS: A DATA BASE The material we gathered lends itself to various analyses. Indeed, within the same framework, we can explore what children, parents and schoolteachers, respectively, think about children’s rights. Through the parents’ and teachers’ questionnaires, we can also get helpful hints about the living conditions of our children and relate them to their views on their rights. What is even more important, is that we can explore which are, if any, the relationships between children’s and parents’ views of children’s rights; furthermore, the same can be done with school teachers, even though more tentatively and speculatively. The scope of this paper, however, is much more restricted. Indeed, we shall confine our analysis to the children’s perceptions of their rights, first at the generational level and then tying them to their parents’ social background. More specifically, at first we intend to give a general and quantitative account of the children’s answers, searching for possible peculiarities linked to their basic characteristics such as country, sex, age and rural/urban milieu. Secondly, we shall explore some possible relationships between the parents’ educational and occupational status and children’s answers. In this way we hope to give a general, yet as much precise as possible frame of reference of the basic results of our research, as to better the direct next steps in fully exploiting its results. For instance, in this paper we shall take into consideration only the “yes/no” answers, without setting about the “why” and other qualitative answers, leaving the task to tackle the “why” matter to a next paper. Before entering upon the data, in the questionnaire addressed to children the word “right” comes into view only in its final section. The dilemmas we proposed to our children were phrased so to avoid any specific and direct reference to the word “right.” We did it purposefully because we believe that our dilemmas clearly imply moral choices, embedding, as we already mentioned, the assertion of the right to do or not to do something for him/herself or for someone else. The equivalence between moral understanding and right reasoning is debatable, not only with regard to children, but to adults too. This is not the place to enter the matter; however, we are sure that they are such close-knit dimensions of any individual and group life. All in all, as suggested by Raes (1997), quite often we resort to rights whenever our moral feelings are not met. We do all of that to justify our constant reference to children’s rights.

136

ANGELO SAPORITI ET AL.

Children’s Answer: The Data Set As we already said, our children were confronted with 16 dilemmas implying a moral choice. All items were phrased to let children say “Yes” or “No” to the solution proposed by the researchers. For the sake of clarity, in Table 2 we report the way in which the dilemmas have been phrased and the proposed solution. Moreover, given that the “Yes/No” answers can assume a completely different meaning depending on the content of the dilemma they refer to, we present only one horn of the quandary, specifying its meaning. Table 2 shows the other three “yes/no” questions about what children think about their rights as well. As we already said, it is not possible to make comments on the children’s perception of their rights with the sole support of the data in Table 1. It would require us to go into every dilemma, clarify and, if possible, classify them in terms of rights; then they should be grouped according to some criterion – for instance, the 3 Ps or “generation” of human rights (Verhellen, 1997a, b, 51ff), and so on. Table 2. Children’s Perception of their Rights (%).a,b 1.

2.

3.

4.

To celebrate the end of school year, your teacher has organized a school recital. She has charged Alina – one of your school-mates – with an important character. Alina is a foreign girl who doesn’t speak Italian very well and she still has problems in making herself understood. For this reason, some of your school-mates fear that the recital will not be properly performed. Therefore, they would like to ask the teacher to exclude Alina from the recital assigning her some other task.c What do you think about that? Should Alina still play the character? Yes: Children asserting the right of Amina to take part in the recital

90.4

Also Ahmed is one of your school-mates. He is a foreign boy as well and everyday he has his meals with you at the refectory. On Tuesday the meal includes pork chops, but Ahmed is a Muslim and he cannot eat this kind of meat. This is the reason why on Tuesday he eats only the first course.c,d What do you think about that? Do you think the school should offer a different second course for everybody to let Ahmed eat the second course, too? Yes: Children who are ready to modify their food habits to let Ahmed share with them the second course You have received a letter from a friend of yours, but you noticed that your parents had opened and read it.c Do you worry about that? Yes: Children worried about their parents’ behavior

51.4

Your parents want you to do a specific activity they believe is well-paying for your well-being and your future, but you really don’t like it. You would rather do something else. What do you think about that? Do you think it is right that your parents insist on persuading you to do what they have already decided? No: Children protesting their right to do what they prefer

73.8

34.7

Children’s Views on Children’s Rights

137

Table 2. (Continued ) 5.

Maria and Paolo are two children who live in your neighborhood. Their parents are going through a difficult period and this prevents them from taking care of both youngsters.c What do you think about that? Do you think it is just that one of the two goes to stay with another family in another town for that period? No: Children asserting Maria and Paolo’s right to remain with their parents

82.0

6.

Antonia is one of your class-mates. Her parents are divorcing and the Judge has to decide on the parent she has to go to live with.c What do you think about that? Do you think the Judge has to decide taking into consideration Antonia’s opinion above all? Yes: Children stating Antonia’s opinion should be given priority by the Judge

83.6

7.

Carles is one of your class-mates. His parents have an opportunity to change their jobs for the better, but they have to move in another town. This means Carles will have to attend another school with other children he doesn’t know, therefore he won’t be able to play and stay with his old friends any more. What do you think about that? Do you think it is fair the whole family moves to another town? No: Children saying that it is not just to move far from their friends and their school

63.1

8.

Coming back from school, you would like to watch a TV program you like very much, but your parents say it doesn’t suit children, so they keep you from watching it. What do you think about that? Do you think it is fair? No: Children saying that it is not just their parents keep them from watching TV programs they like Rwanda is an African poor country. It is wanting teachers, and its Government has asked for support to the international community, Italy included. Therefore, your teacher has decided to take a six months leave to go to teach in Rwanda.c What do you think about that? Do you think it is fair your teacher leaves you and your friends for such a long time? Yes: Children stating it is just their school teacher moves to a poor country to educate less fortunate children You and your friends have decided to organize a running race for your district’s festival. The leaders of the celebration agree with you but they have decided it is up to your parents to organize the running race. What do you think about that? Do you think it is fair? No: Children stating their right to organize and manage themselves their activities

9.

10.

52.8

85.5

57.1

11.

Marta always spends weekends at home with her parents, while her friends go on a Sunday trip. Marta would like to go out on Sunday too, but her parents want her at home with them. What do you think about that? Do you think it is fair? No: Children saying they would prefer to spend weekends together with their friends

88.4

12.

Now and again Jos´e – one of your class-mates – skips school because he must help his parents to run their shop.c What do you think about that? Do you think it is fair? No: Children stating Jos´e’s right to attend school

89.6

138

ANGELO SAPORITI ET AL.

Table 2. (Continued ) 13.

14.

15.

16

17. 19. 20.

In your school there are some misbehaving pupils who disturb their mates and the lessons. Some teachers would rather send them away from the school to prevent harming the whole class. There is a teacher, however, who says that their class-mates should help them have a proper conduct. What do you think about that? Do you think the teacher is right? Yes: Children who agree with the teacher saying that it is up to them to help their misbehaving mates In´es is a sightless girl attending a school for blind pupils. However, she would like to go to a normal school with her friends.c What do you think about that? Do you think In´es should attend a school including both sightless and sighted children? Yes: Children saying In´es should share a normal class-room with her friends

70.3

Gianni’s parents are deciding where to spend the summer holidays. What do you think about that? Do you think they should ask Gianni where he would like to go before taking a decision? Yes: Children asserting Gianni’s right to have a voice about his summer holidays

93.4

81.9

Carlo is not very good at school and he may fail to pass. His parents told him everybody should manage his/her school problems on his/her own. What do you think about that? Do you think Carlo’s parents are right? No: Children saying they should be supported by their parents in facing school problems Do you know what a right is? Yes: Children saying they know what a right is

92.1

Do children have rights? Yes: Children saying that children do have rights

97.0

Do you know which the rights of the children are? Yes: Children saying they know which the rights of the children are

81.1

71.3

a For

the complete data set see table Appendix. item number corresponds to the one in the questionnaire. c“ Why” questions. d Dilemma 2 has been worded in a different way in the two languages. b The

This is not our task in this paper. We prefer to follow another design. Is that set of answers homogeneous, i.e. invariant with respect to some basic factors as country, sex, age, and residential area (country or town)? Molise vs. Catalonia Let us start by saying that 11 of 19 questions listed in Table 1 got the same answer from more than 80% of our children. Therefore, it will not be possible for them to find factors able to produce clear and telling differences.

Children’s Views on Children’s Rights

139

Table 3. Children’s Perception of their Rights, by Country (%). Items 2. Children who are ready to modify their food habits to let Ahmed share with them the second course c 4. Children protesting their right to do what they prefer 8. Children saying that it is not just their parents keep them from watching TV programs they like 9. Children stating it is just their school teacher moves to a poor country to educated less fortunate children 11. Children saying they would prefer to spend weekends together with their friends 14. Children saying In´es should share a normal class-room with her friends 16. Children saying they should be supported by their parents in facing school problems 17. Children saying they know what a right is 19. Children saying that children do have rights 20. Children saying they know which the rights of the children are

Cab

␹2

45.3

19.0

116.92

76.9 43.2

69.3 67.1

11.51 86.98

83.6

88.3

6.80

90.4

85.3

9.81

66.4

75.9

16.57

78.1

61.1

54.17

94.3 98.4 89.0

88.7 94.8 69.0

16.76 17.60 98.52

Moa

Note: Significant Differences (␣ ≤ 0.01). a Molise. b Catalonia. c Dilemma 2 has been worded in a different way in the two languages.

Having said that, we start with the countries’ differences because we are interested in any difference that could be ascribed to more general social-cultural factors. We prefer to be under strict statistical constraints in accepting possible differences – to put it in technical terms, we shall consider an alpha [␣] of 0.01 as the cutoff for statistical significance. Table 3 displays only the items showing statistically significant differences between the countries. Following the same format of Table 2, it shows only the specified answer; in addition, it shows the ␹2 value. From data displayed in Table 3 we can say that children from Catalonia and Molise disagree on a little more than half the items (10 out of 19). However, putting aside items 17, 19 and 20, we can claim that children from both countries seem to be quite close in their moral orientations. Only 7 dilemmas of 16 are statistically different. Someone could object that this is a rather big difference and that, on the contrary, the two groups of children do have different orientations on the issues presented. In our opinion, however, this is not the case. A closer look at Table 3 tells us that although statistically significant, four of them are not so weighty from a substantial point of view. (The same holds for two of the last three items.) Indeed, only four of the seven differences are really relevant, including dilemma 2 that

140

ANGELO SAPORITI ET AL.

has been worded in a different way in the two languages. Second, the registered differences are not consonant with a homogeneous pattern. Only three dilemmas highlight a clear difference between Catalan and Molisan children: Children saying that it is not just that their parents keep them from watching the TV programs they like, children saying In´es should share a normal class-room with her friends, and children saying they should be supported by their parents at school. In addition, it is worthy to notice the difference (20%) concerning item 20. Gender Differences With gender differences, we have lessened the statistical requirement considering an ␣ of 0.05 as the cutoff for significance (we did the same also with regard to age and residential area variables). Table 4 shows a substantial homogeneity of thoughts among boys and girls. Of all items, only five of nineteen show a (statistically significant) difference but just one is close to 9%, whereas all others are around 5% or a little more. We then took up in commenting country differences, to maintain that gender has little or nothing to do with our children’s choices. Things get a little different, at least as far as Catalonia is concerned, if we cross children’s answers by gender within each country sample (Table 5). While children from Molise do not show substantial differences, those from Catalonia are more gender distinguished, as both to the number of differences and to their size. However, they are not consistent in their direction. Thus, we can specify our previous conclusion saying that gender differences in Catalonia are worth taking into consideration, even though they should not be considered determinant with regard to our children’s views. Age differences (Table 6) are present in only 7 items of 19. Furthermore, only 4 of them (items 2, 8, 9, and 20) show a clear trend (positive or negative). It is worth noting the peculiar trend of item 20: The older children seem to ignore their rights much more than the younger children. Table 4. Children’s Perception of their Rights, by Gender (%). Items

Boys

Girls

␹2

1. Children asserting the right of Amina to take part in the recital 6. Children stating Antonia’s opinion should be given priority by the Judge 12. Children stating Jos´e’s right to attend school 16. Children saying they should be supported by their parents in facing school problems 20. Children saying they know which the rights of children are

87.4 81.6 87.0 67.8

93.5 85.5 92.2 74.9

17.62 4.35 11.59 10.12

76.6

85.5

19.93

Note: Significant Differences (␣ ≤ 0.05).

Children’s Views on Children’s Rights

141

Table 5. Children’s Perception of their Rights, by Country and Gender (%). Items

Country

Boys

Girls

␹2

1. Children asserting the right of Amina to take part in the recital 2. Children who are ready to modify their food habits to let Ahmed share with them the second coursea 4. Children protesting their right to do what they prefer 8. Children saying that it is not just that their parents keep them from watching the TV programs they like 10. Children stating their right to organize and manage themselves their activities 12. Children stating Jos´e’s right to attend school 13. Children who agree with the teacher saying that it is up to them to help their misbehaving mates 14. Children saying In´es should share a normal class-room with her friends 16. Children saying they should be supported by their parents in facing school problems 17. Children saying they know what a right is 19. Children saying that children do have rights 20. Children saying they know which the rights of children are

Mo Ca Ca

87.2 87.6 22.6

92.6 94.8 15.5

7.75 10.55 5.20

Mo

26.6

19.6

6.65

Ca

27.7

38.0

7.64

Ca

40.9

49.7

4.99

Ca Ca

15.9 80.2

6.4 86.6

14.89 4.74

Ca

70.7

81.0

9.25

Ca

44.7

33.3

8.66

Mo Mo Mo Ca

92.6 97.5 86.6 60.9

96.2 99.4 91.5 76.6

5.81 5.21 5.89 17.75

Note: Significant Differences (␣ ≤ 0.05). a Dilemma 2 has been worded in a different way in the two languages.

Once again, if we cross the items by age within each of the two regions (Table 7), things do not change much. While in Molise the number of the significant differences decreases up to 5, in Catalonia it increases by one item only. Therefore, we can say that not even age is a generalized predictor of our children’s views on their rights. However, the fact that the differences we found are almost all strong and clear enough gives support to the hypothesis that age differences must be given some place in the explanation of children’s understanding of their rights. Age Differences Residential Differences If neither country nor gender and age are strongly related to the ways our children faced the dilemmas they were confronted with, the rural/urban milieu has nothing

142

ANGELO SAPORITI ET AL.

Table 6. Children’s Perception of their Rights, by Age (%). 10

11

12

13

␹2

49.5

39.1

34.7

30.8

21.5

23.14

54.3

49.5

47.1

55.4

63.2

13.84

38.0

46.8

52.5

57.7

68.6.

28.26

79.3

81.6

86.2

87.0

94.4

15.58

90.3

84.3

90.2

91.0

81.9

15.68

77.2

85.0

84.1

80.4

67.3

21.97

85.7

86.0

81.2

78.3

69.2

19.00

Items

9

2. Children who are ready to modify their food habits to let Ahmed share the second course with thema 3. Children worried about their parents’ behavior 8. Children saying that it is not just that their parents keep them from watching the TV programs they like 9. Children stating it is just their school teacher moves to a poor country to educated less fortunate children 11. Children saying they would prefer to spend weekends together with their friends 13. Children who agree with the teacher saying that it is up to them to help their misbehaving mates 20. Children saying they know which the rights of children are

Note: Significant Differences (␣ ≤ 0.05). a Dilemma 2 has been worded in a different way in the two languages.

to do with the matter, neither generally nor within each country. Tables 8 and 9 show this fact very clearly, without needing any comment.

Concluding Remarks Social scientists are always in search of factors that “make the difference” and they are generally disappointed in discovering uniformity of attitudes or behaviors. This was our case, too. Of course, someone may have the opposite opinion – i.e. that we found quite a lot of differences among our children that can be attributed to the factors we introduced, but this is not our thought. Certainly, there are dissimilarities between children from Molise and Catalonia, boys and girls, as well as among children of different ages. These differences are worthy to be deepened, but they are no more than that. Maybe, only age makes a difference that deserves being considered relevant.

Children’s Views on Children’s Rights

143

Table 7. Children’s Perception of their Rights, by Age and Country (%). 10

11

12

13

␹2

41.7

41.9

56.8

60.0

16.26

72.1

79.9

82.1

80.0

77.5

86.8

36.7

45.0

50.0

Ca

83.5

87.5

88.5

95.7

7.55

Ca

48.7

48.4

57.3

70.3

14.73

Ca

78.3

87.4

88.5

81.3

8.24

Ca

92.1

85.3

83.6

68.1

4.63

Ca

64.3

78.7

79.0

76.7

10.51

78.5 50.5

79.1 58.1

83.2 66.7

68.1

13.49 10.89

Items

a

3. Children worried about their parents’ behavior 4. Children protesting their right to do what they prefer 6. Children stating Antonia’s opinion should be given priority by the Judge 8. Children saying that it is not just their parents keep them from watching TV programs they like 9. Children stating it is just their school teacher moves to a poor country to educate less fortunate children 10. Children stating their right to organize and manage themselves their activities 11. Children saying they would prefer to spend weekends together with their friends 13. Children who agree with the teacher saying that it is up to them to help their misbehaving mates 14. Children saying In´es should share a normal class-room with her friends 16. Children saying they should be supported by their parents in facing school problems 17. Children saying they know what a right is 19. Children saying that children do have rights

Ca Mo

9

68.8

Ca

Mo

Mo Ca Mo Mo

38.0

64.5

87, 0 96.7

95,7 96.8

96,1 99.7

94,1 99.0

12.10 90.2

10.91

10.15

12.933 10.84

Note: Significant Differences (␣ ≤ 0.05). a Country.

Surely, this is only a prima facie conclusion, and we shall put off exploring the issue to another paper introducing other variables. However, before we come to some general conclusion, let us try to discover other factors that may make more substantial differences than the ones we found. Thus we shall introduce the “parents” variable.

144

ANGELO SAPORITI ET AL.

Table 8. Children’s Perception of their Rights, by Residential Area (%). Items

Rural Areas

␹2

49.4 82.3

55.3 86.1

4.97 3.83

93.1

90.1

4.48

Urban Areas

3. Children worried about their parents’ behavior 6. Children stating Antonia’s opinion should be given priority by the Judge 17. Children saying they know what a right is Note: Significant Differences (␣ ≤ 0.05).

INTRODUCING PARENTS’ SOCIAL AND CULTURAL BACKGROUND The parents’ questionnaire includes several questions focused on analyzing both the social context in which children live and the parents’ attitude towards their children and some of their rights. The way the data have been gathered is such to allow us to construct a child-parent relationship in order to analyze whether some of the parents’ basic characteristics and attitudes toward their children show some Table 9. Children’s Perception of their Rights, by Country and Residential Area (%). Country

3. Children worried about their parents’ behavior 6. Children stating Antonia’s opinion should be given priority by the Judge 8. Children saying that it is not just their parents keep them from watching TV programs they like 12. Children stating Jos´e’s right to attend school 14. Children saying In´es should share a normal class-room with her friends 17. Children saying they know what a right is 20. Children saying they know which the rights of children are

CA

45.1

58.5

9.96

CA

80.1

88.6

7.25

MO CA

46.0 63.8

37.8 73.7

5.79 6.14

CA

91.5

83.8

8.45

MO

64.1

71.0

4.60

CA

91.1

83.8

7.41

MO CA

86.6 72.3

93.7 62.1

10.88 6.71

Note: Significant Differences (␣ ≤ 0.05).

Urban Area

Rural Area

␹2

Items

Children’s Views on Children’s Rights

145

kind of relation with children’s perception of their rights. However, given that this is a task far beyond the scope of this paper, we shall only take into consideration parents’ social and cultural background, leaving aside their attitudes towards their children’s rights. Common sense and scientific literature suggest that cultural and economic conditions exert a certain weight in determining choices and behavior, both directly and obliquely. Thus we can hypothesize that our children can be differentiated as to their solution to the dilemmas we proposed, according to their parents’ education and economic situation – the latter being measured by the proxy “occupational status” or job. In order to explore whether the parental status is linked to the children’s answers, we have at first elaborated two indexes to synthesize their educational and occupational status, plus a general index to assess the overall family status, summing up the former indexes.13 As we have already said, the social condition of both samples is a middle-class situation. However, even though we purposefully avoided taking into consideration the extreme situations from both sides of the social scale (lower and upper class), we can distribute our parents along a three portioned continuum going from a lower-middle class status to an upper-middle class status. Of course, there are no sharp qualitative differences between the three groups, but according to our knowledge the differentiation we got is quite representative of the situation in both regions. Table 10 displays the data on social status. To analyze the effects of the parents’ social status on their children’s perception of rights, we have used the same blueprint we adopted to check the effects of sex, age and residential area – i.e. cross tabulation between the children’s answers and the parents’ occupation and level of education, as well as the family social standing. In this case, too, we used ␣ = 0.05 as the statistical significance indicator.

Table 10. Parents’ Social Standing, by Country (%). Middle Class Levels

Lower Middle Upper Total #

Education

Occupation

Family Status

Molise

Catalonia

Molise

Catalonia

Molise

Catalonia

37.2 48.7 14.0

19.1 42.3 38.6

49.3 41.0 9.7

19.0 63.3 17.7

59.9 23.5 16.5

24.4 46.6 29.0

100.0 835

100.0 241

100.0 839

100.0 368

100.0 841

100.0 369

146

ANGELO SAPORITI ET AL.

Table 11. Children’s Perception of their Rights, by Parents’ Social Standing and Country. Item

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 19. 20.

Education

Occupation

Region

␹2

␣

Mo

8.07

0.018

Mo

6.27

0.043

Ca Mo

12.87 14.37

0.002 0.001

Ca Mo

6.67 8.36

0.036 0.015

Mo Mo

8.96 20.18

0.011 0.000

Family Status

␹2

␣

Mo

15.72

0.000

Mo Ca

8.61 7.38

Mo

12.74

Region

␹2

␣

Ca Mo

8.47 26.24

0.014 0.000

0.013 0.025

Mo

13.12

0.001

0.002

Mo Ca

16.23 11.92

0.000 0.003

Region

Note: Social Standing Significant Differences within Countries (␣ ≤ 0.05).

Previous findings concerning the effects of children’s basic features (gender, sex, age and residential areas) didn’t make us skeptical about finding differences to be charged to the parents’ different background. Table 11 shows clearly that more than in previous cases the children’s answers were not related to their parents’ social status. Only 6 out of 19 items were related to the parents’ educational level in Molise, and hardly 2 in Catalonia, let alone occupational and family status, where the number of significant differences is beside the point. In our opinion the solution displayed in Table 10 is the best to differentiate the parents’ social standing. Nonetheless, we computed other arrangements of the parents’ background variables, but none was more discriminating than the one showed in Table 10. The same holds true when comparing only the two extremes (lower and upper) of the parents’ social standing continuum. Parents’ education seems to be a little more linked to the children’s answers than the other two variables.

Children’s Views on Children’s Rights

147

CONCLUDING REMARKS: IS CHILDREN’S REASONING ABOUT THEIR RIGHTS REALLY A “FLAT WORLD?” The previous sections show some studies on children’s views of their rights have been reviewed. All of them show that the idea that children reason about their rights is swayed to some extent by factors such as age, gender, culture, as well as social context and living condition. Even though it is not yet clear whether these factors are arranged according to some hierarchical order, there is evidence of their impact on children’s perception of their rights. Together with all the cautions we have previously expressed, our research provides evidence against those results, even though we have not yet explored all the potentialities of the available data. It seems as if our children lived in a “flat world” where nothing makes differences. Is that true? Maybe we are underrating the differences we found. If not, is there something wrong with our study? We do not have ready-made answers to these interrogatives. However, we can put forward some tentative hypothesis together with some suggestions for future research and analysis. First of all, from a methodological point of view it is something like a postulate that no constant or uniformity can make differences. This applies to both children and parents. As to the latter case, perhaps they are too much akin to “produce” weighty differences on their children. Should it be true, future research should address samples much more differentiated, especially as far as social and living conditions are considered – including, for instance, lower and upper classes and/or people living in big cities. On the other hand, the fact that many dilemmas and other questions included in the children’s questionnaire received the same answer by the great majority of the children leaves not much scope for searching variations. There are many other wellsuited questions to be analyzed from our point of view. Thus, we can wonder if the questions we posed to our children are really “non-discriminating” with regard to their moral reasoning. An answer to such a question requires that the homogeneity we found applies to more varied situations. So, the research suggestion is the same as the previous one, plus the more trivial reported in the next point. A second more obvious hypothesis suggests that we have not yet explored all the information we gathered. For instance, we have not taken into consideration the relationships between parents’ attitudes toward children’s rights and their children’s view on the same subject matter. Furthermore, we have not analyzed the role of school, neither in general nor with regard to schoolteachers. In the same way, in this paper we have not taken into account other social factors belonging to the social context in which our children are living. It goes without saying that this is our main task in the short run.

148

ANGELO SAPORITI ET AL.

Third, differences could be less relevant than we believe, as entering the meaning of the dilemmas and trying to give a proper sense to the children’s answers is much more important. That’s a job we have already started and it is a very promising one. Finally, there is a much more intriguing hypothesis than all the previous ones. Our Western, “post-industrialized” and “global village” children might develop and acquire a moral and right sense under circumstances that have to do more with peers and media than with the traditional institutions of socialization, such as family and school. No human being develops his/her cognitive and judgmental capacities in a vacuum. In a world in which the media seem to be a cultural and social leveler, and the peer group a socialization channel outweighing the family and the school as the source of values and behavioral patterns, to speculate on an ongoing process of homogenization of children’s moral and rights reasoning is worthy of belief. Surely it goes against current evidence, but we have to remark that most of this evidence concerns legal and political evidence – a field which is extremely relevant but highly complex and technical and, above all, just a piece of the children’s rights world. We hope our research and our findings will add another small brick to our understanding of children’s views of their own rights – first of all for the children themselves, but for us, too.

NOTES 1. See, e.g. the historical record by Veerman (1992) and, as far as the U.S. are considered, Bottoms, Bull Kovera and McAuliff (2002). See also Saulle (2003). 2. For instance, in the way Holt (1974) puts it. 3. “Children’s rights” is a general term covering children adolescents and youths who have not yet reached the age of maturity according to the State legislation. However, in this paper it refers mainly to children and young adolescents. 4. The literature on children’s rights is by now broad enough to show the pros and cons of children’s rights, the various positions on the subject, as well as the many facets it has. For sake of brevity, we shall avoid reporting such literature; it will be sufficient to mention two opposing points of view: Purdy (1992) has a negative view, while Franklin (1992) is in favor. Freeman (1983), even being favorable, presents a more balanced and articulated position. 5. This section, as well as the following one, is strongly based on Casas, Saporiti et al. (forthcoming). 6. On this point see, among others: Qvortrup (2001); Qvortrup et al. (1994); Verhellen (1997); Casas (1998); Saporiti (2001). 7. See the review by Schmidt and Reppucci (2002). 8. For an extensive analysis of Piaget’s and Kohlberg’s works, see Hallpike (2002).

Children’s Views on Children’s Rights

149

9. Small and Limber (2002), as well as Schmidt and Reppucci (2002), offer an extensive and updated review of empirical research on children’s perception and understanding of legal concepts and rights. 10. In this connection it’s worthy to mention the work by Taylor, Smith and Nairn (2001), who have analyzed a wide sample of 821 youths from New Zealand and a sample of 438 education professionals of the same school – note, however, the absence of the youths’ parents. 11. The questionnaires for parents and teachers followed a similar construction as the children’s questionnaire. 12. The 16 children from Molise aged thirteen have been excluded from any agecomparison between the two regions. 13. To build these indexes we have used the data set on education and occupation from the parents’ returned questionnaire. However, in order to have a sample size as large as possible we have also: (a) used the same information from the children’s questionnaires whenever the parents’ ones were lacking; and (b) whenever the educational/occupational status of one of the two parents was missing, we have attributed him/her the educational/occupational status of the other. Thus, we have excluded from the analysis only those returned questionnaires missing both the parents’ educational/occupational status. Finally, we have also combined the two indexes in order to have an index referring to the family as a whole.

REFERENCES Archard, D. (1993). Children: Rights and childhood. London and New York: Routledge. Aristotele, E. N. (1999). Introduzione. Note and translation of C. Natali. Milano: Mondolibri S.p. A., su licenza Gius. Laterza and Figli. Backe-Hansen, E. (2003). Cool, boring, difficult, stupid? What the children thought Methods and techniques in child research in international comparison. M¨unich: Deutsches Jugendinstitut (DJI). Blustein, J. (1982). Parents and children The ethics of the family. Oxford: Oxford University Press. Casas, F. (1997a). Children’s rights and children’s quality of life: Conceptual and practical issues. Social Indicators Research, 283–298. Casas, F. (1997b). Quality of life and the life experience of children. In: E. Verhellen (Ed.), Understanding Children’s Rights. Collected papers presented at the Second International Interdisciplinary Course on Children’s Rights, July 1997. Belgium: Children’s Rights Centre, University of Ghent. Casas, F. (1998). Infancia: Perspectivas psicosociales. Barcelona: Paid´os. Casas, F., & Saporiti, A. et al. (forthcoming). Los derechos de la infancia desde la perspectiva de los propios ni˜nos y ni˜nas, de sus progenitores y de sus maestros: Un estudio comparativo entre Catalu˜na (Espa˜na) y Molise (Italia). Chawla, L. (2001). Evaluating children’s participation: Seeking areas of consensus. In: IIED, Children’s Participation – Evaluating Effectiveness. Pla notes. Participatory learning and action. London: International Institute for Environment and Development (IIED). Cherney, I., & Perry, N. W. (1994). Children’s attitudes toward their rights. In: F. Casas (Ed.), Psychological Perspectives on Childhood. Symposium in the 23rd International Congress of Applied Psychology, Madrid.

150

ANGELO SAPORITI ET AL.

Clark, S. R. L. (1989). Children and the mammalian order. In: G. Scarre (Ed.), Children, Parents and Politics. Cambridge: Cambridge University Press. Franklin, B. (1992). Votes for children. Childright, 88, 10–14. Freeman, M. (1983). The rights and wrongs of children. Pinter. Hallpike, C. R. (2002). Evolution of moral understanding. Prometheus Research Group. Holt, J. (1974). Escape from childhood: The needs and rights of children. New York: Ballantine. Kohlberg, D. (1981). The philosophy of moral development. Moral stages and the idea of justice. Essays on moral development (Vol. 1). San Francisco: Harper and Row. Mill, J. S. (1910). On liberty. London: Dent (Everyman series). Melton, G. B. (1980). Children’s concepts of their rights. Journal of Clinical Child Psychology, 9, 186–190. Melton, G. B. (1983). Child advocacy: Psychological issues and intervention. New York: Plenum Press. Melton, G. B., & Limber, S. (1992). What children’s rights mean to children: Children’s own views. In: M. Freeman & P. Veerman (Eds), Ideologies of Children’s Rights. Dordrecht, Netherlands: Martinus Nijhoff. Ocha´ıta, E., Espinosa, M. A., & Grediaga, M. C. (1994). ?C´omo entienden los ni˜nos el derecho a la igualdad? Infancia y Sociedad, 27/28, 61–76. O’Neill, O. (1996). Children’s rights and children’s lives. In: R. Ekman Ladd (Ed.), Children’s Rights Re-Visioned. Wadsworth Publishing Company. Platone, La Repubblica, R. C. S. Libri, & Grandi Opere S.p. A., 1994, Volume 2 (Introduzione di F. Adorno, trad. di. F. Gabrielli). Piaget, J. (1932). The moral judgment of the child (Trans M. Gabain). London: Routledge and Kegan Paul. Purdy, L. M. (1992). In their best interest? The case against equal rights for children. Ithaca and London: Cornell University Press. Qvortrup, J. (2001). Childhood as a social phenomenon revisited. In: M. Du Bois-Reymond, H. S¨unker & H. H. Kr¨uger (Eds), Childhood in Europe. Approaches−Trends−Findings. New York: Peter Lang Publishing. Qvortrup, J. et al. (Eds) (1994). Childhood matters, Social theory, practice and politics. Vienna: Aldershot, Avebury and European Centre for Social Welfare Policy and Research. Raes, K. (1997). Children’s rights and modern culture. The politics of juridification and its limits. In: E. Verhellen (Ed.), Understanding Children’s Rights. Collected papers presented at the Second International Interdisciplinary Course on Children’s Rights, July 1997. Belgium: Children’s Rights Centre, University of Ghent. Ruck, M. M. et al. (1998). Adolescents’ and children’s knowledge about rights: Some evidence for how young people view rights in their own lives. Journal of Adolescence, 21, 275–289. Saporiti, A. (2001). A methodology for making children count. In: M. Du Bois-Reymond, H. S¨unker & H.-H. Kr¨uger (Eds), Childhood in Europe. Approaches−Trends−Findings. New York: Peter Lang Publishing. Saulle, M. R. (2003). Le norme internazionali a tutela dei minori nella Convenzione sui diritti del fanciullo. In: D. Grignoli, A. Mancini & A. Saporiti (a cura di), Infanzia, educazione ai diritti umani e benessere sociale. Napoli: ESI. Schmidt, M. G., & Reppucci, N. D. (2002). Children’s rights and capacities. In: B. L. Bottoms, M. Bull Kovera & B. D. McAuliff (Eds), Children Social Science, and the Law. Cambridge: Cambridge University Press.

Children’s Views on Children’s Rights

151

Small, M. A., & Limber, S. P. (2002). Advocacy for children’s rights. In: B. L. Bottoms, M. Bull Kovera & B. D. McAuliff (Eds), Children Social Science, and the Law. Cambridge: Cambridge University Press. Sparks, R., Girling, E., & Smith, M. (2000). Children talking about justice and punishment. The International Journal of Children’s Rights, 8(3), 191–209. Tapp, J. L., & Levine, F. J. (1974). Legal socialization: Strategies for an ethical legality. Stanford Law Review, 27, 1–72. Taylor, N., Smith, A. B., & Nairn, K. (2001). Rights important to young people: Secondary student and staff perspectives. The International Journal of Children’s Rights, 9, 137–156. Veerman, P. E. (1992). The rights of the child and the changing image of childhood. Dordrecht/Boston/London: Martinus Nijhoff Publishers. Verhellen, E. (Ed.) (1997). Understanding children’s rights. In Collected papers presented at the Second International Interdisciplinary Course on Children’s Rights July 1997. Belgium: Children’s Rights Centre, University of Ghent. Verhellen, E. (1997). Convention on the rights of the child (2nd ed.). Leuven/Apeldoorn: Garant. Walker, S. (2001). Consulting with children and young people. The International Journal of Children’s Rights, 9, 45–56.

APPENDIX: CHILDREN’S DATA SET (#)

Question

Answers

Q. 1

Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No

Q. 2 Q. 3 Q. 4 Q. 5 Q. 6 Q. 7 Q. 8 Q. 9

All 1,449 153 556 1,045 819 774 420 1,185 288 1,315 1,331 262 590 1,010 751 839 1,367 232

Molise 863 97 434 525 505 453 223 741 175 787 797 152 378 584 542 413 801 157

Catalonia 586 56 122 520 314 321 197 444 113 528 534 110 212 426 209 426 566 75

152

Q. 10 Q. 11 Q. 12 Q. 13 Q. 14 Q. 15 Q. 16 Q. 17 Q. 19 Q. 20

ANGELO SAPORITI ET AL.

Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No

684 909 186 1,413 167 1,433 1,308 290 1,117 473 1,502 106 459 1,143 1,464 96 1,543 48 1,273 297

394 560 92 868 96 861 773 184 631 319 901 64 211 753 900 54 946 15 844 104

290 349 94 545 71 572 535 106 486 154 601 42 248 390 564 72 597 33 429 193

INVOLVING CHILDREN IN SOCIAL POLICY: A CASE STUDY FROM NORTHERN IRELAND Madeleine Leonard INTRODUCTION One of the most notable breakthroughs in promoting the right of children to be consulted about policies that affect them is the UNCRC (United Nations Convention on the Rights of the Child, General Assembly of the United Nations, 1989). At the time of writing, the UNCRC has been ratified by all but two (Somalia and the USA) member states of United Nations. The Convention was ratified by the United Kingdom in 1991 and according to Daniel and Ivatts (1998, p. 16) “it is arguably the most significant development in United Kingdom policy towards children since 1945.” By ratifying the Convention, governments must take steps to ensure that they meet the standards and principles set out in the various Articles in the Convention and must provide regular reports to the UN Committee on the Rights of the Child on its implementation. One of the most significant Articles is Article 12 which specifies that: State Parties shall assure to the child who is capable of forming his or her own views the right to express those views freely in all matters affecting the child, the views of the child being given due weight in accordance with the age and maturity of the child (General Assembly of the United Nations, 1989).

Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 153–167 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10008-1

153

154

MADELEINE LEONARD

This Article has found its way into United Kingdom social policy through the introduction of the Children’s Act in 1989 (this Act was extended to Northern Ireland in 1995). This provides children in care or children whose parents are going through a divorce some involvement in the decision-making process. The Children’s Rights Development Unit which monitors the United Kingdom implementation of the Convention has made Article 12 the primary focus of its work (Shier, 2001). This is partly due to the reaction of the UN Committee on the Rights of the Child (1995) to the United Kingdom’s first report. The Committee identified a number of gaps in the implementation of children’s rights particularly in the area of participation rights. Indeed, children’s voices remain largely absent in many areas of social policy particularly those relating to education and health (Daniel & Ivatts, 1998). As Hill and Tisdall (1997, p. 256) put it “the rhetoric of children’s participation is easier and cheaper than its effective implementation.” Involving children in policy throws up all sorts of issues relating to the participation of children in the decision-making process. What does adopting a child-centred approach to children’s role in decision-making entail? How do we ensure that children’s views are effectively incorporated in the policy-making arena? How do we find out about the views and perceptions of children in relation to whatever issue is being debated? As Hill and Tisdall (1997) point out, involving children in social policy often gets narrowly translated to listening sympathetically to their views rather than considering them as a social group capable of influencing policy and practice. Yet organisations, which promote the rights of children, such as Save the Children, argue that more meaningful social policies will evolve from taking on board the perspectives of those who are influenced by such policies. One useful model that could be employed to ensure that children effectively participate in research linked to social policy is Hart’s ladder of participation (1992). A survey of children’s organisations throughout the United Kingdom in their attempts to introduce mechanisms to ensure that their policies and decisions take more account of children’s opinions revealed that Hart’s model played a significant role in their strategies to involve children in policy related research (Barn & Franklin, 1996). Academic researchers have also utilised Hart’s ladder as a framework for advocating approaches to enhance the participation of children in the research process (Landsdown, 1995; Shier, 2001; Verhellen, 1997). The purpose of this paper is to examine the usefulness of Hart’s model in involving children in social policy related research concerning the Eleven Plus system in Northern Ireland.

HART’S LADDER OF PARTICIPATION Hart’s (1992) intention was to provide a framework for participation, which would facilitate the involvement of children in research linked to social policy. His model

Involving Children in Social Policy

155

outlines an eight-ring ladder of participation and draws on a similar eight-rung ladder put forward by Arnstein (1969) to encourage citizen participation in policy. The first three rungs are labelled: Manipulation, Decoration and Tokenism. Hart uses these terms to describe research where children have no real understanding of the issues being researched. Or they may understand the issues but have no control over how they express their views. Additionally adults may use some of the ideas expressed by children but do not tell them what influence they had on the final decision or how their views are being used. According to Shier (2001) one of the most useful benefits of Hart’s ladder has been to identify these superficial examples of children’s participation. To some extent, these issues reflect research on children until the last two decades. Brannen and O’Brien (1995) suggest that researchers often carried out research on children rather than with children. This reflected a tendency to see children as incompetent, immature and incapable of actively taking part in research. Hence, research was done on children with adults positioned as the experts on children’s lives. Indeed, some studies supposedly aimed at capturing children’s worlds interviewed adults as parents to find out about everyday aspects of children’s lives (Gogle & Tasker, 1982; Goodnow & Delaney, 1989; White & Brinkerhoff, 1981). This reflects Lansdown’s (1994, p. 38) observation that we do not have “a culture of listening to children.” Moreover, much of the research involving children tended to see them as “human becomings” rather than as human beings (Qvortrup, 1987). This future-orientated focus rendered invisible many aspects of children’s everyday lives. Hart (1992) labels the fourth rung of his ladder as “Assigned but Informed”. This refers to research where adults decide on the project but take active steps to ensure that children are fully informed about the research and why they should be involved. The fifth rung is entitled “Consulted and Informed.” Here, children are consulted about the research and their opinions are taken seriously. “Adult-Initiated, Shared Decisions with Children” is the sixth rung of the model. This is where children are involved at various stages of the research process. Collectively, these three rungs represent strategies to encourage the active participation of children in the research process. There have been many successful attempts to incorporate more active input from children in the research process based on Hart’s model (Lansdown, 1995; Milner & Carolin, 1999; Verhellen, 1997). This research highlights the necessity of taking children’s views and perspectives into account in formulating policies, which impact on children’s lives. The final top two rungs of the model represent child-oriented research and Hart admits that this is the least likely form of children’s participation. These two rungs of Hart’s ladder refer to incidences where research is initiated and carried out by children. Adults may be involved in such projects but children remain the primary decision-makers. Hart implies that child-initiated research is potentially superior to all other forms of research. However, there are a number of problems associated

156

MADELEINE LEONARD

with children carrying out their own research, which the academic community has thus far failed to address. Some researchers have involved children in research as co-researchers (see Alderson, 1995, 2001) but such research continues to be written up by adults. Therefore even when children play such an active role by researching each other, the power to interpret and analyse what they say remains largely in adult hands. Indeed, Harden et al. (2000) suggest, that incorporating children into the research as researchers may create new levels of hierarchy. Children may find themselves in such circumstances in a limbo area positioned between adult researchers and other children who are respondents and with the shared status of childhood being used by adults to gain access to children’s worlds. Moreover, children are rarely involved in research at the proposal stage. To obtain funding, proposals have to be written in such a way that they appeal to adult evaluators. As Pole, Mizen and Bolton (1999, p. 50) point out “the prospect of giving out large amounts of public money to children to conduct their own research is likely to halt even the most committed child-centred researchers in their tracks.” Overall, Hart (1992) sub-divides his ladder into three groups. The first three rungs represent ineffective ways of involving children in the research process. The second three rungs represent effective ways of involving children in research while the final two rungs represent research initiated and carried out by children. The research on which this paper is based can be placed in Hart’s middle grouping referring to effective ways of incorporating children into the research process and ensuring that relevant policy-makers hear their voices.

THE ELEVEN PLUS SYSTEM IN NORTHERN IRELAND Before discussing the application of Hart’s model, it is necessary to set the context for the paper by briefly presenting the background to the Eleven Plus system in Northern Ireland and the current debates that are taking place regarding the future of the exam. The Eleven Plus was introduced throughout the United Kingdom in 1944 with the primary aim of distinguishing between children perceived to have an academic ability and children perceived to have a vocational ability. Based on the test, children were sent to separate types of schools geared towards developing these different abilities and the system operated as a sorting device for preparing children for their future roles in the labour market. Hence, the test mirrored the common tendency to regard children in terms of what they would become, a perspective which rendered invisible everyday aspects of their daily current lives. Children who passed the test were sent to grammar schools while children who failed the test were sent to secondary schools. Social class emerged as the most likely variable affecting performance in the test and gradually grammar

Involving Children in Social Policy

157

schools came to represent middle class institutions while secondary schools came to represent working class institutions (Gallagher & Smith, 2000; Leonard & Davey, 2001). Secondary schools also became regarded as “second best” and the fact that pupils had to fail a test to gain entry to secondary schools meant that many secondary school pupils started their second level stage of education with very low self-esteem and a sense of failure (Leonard & Davey, 2001; Smith et al., 2000). In a wave of criticism that emerged throughout the 1960s, the exam was abolished in all regions of the United Kingdom except Northern Ireland and replaced with a comprehensive system of education, which brought together children of different abilities within the same school. This system was viewed as a fairer more equitable educational system as it catered to children of all academic abilities. Northern Ireland’s political situation has dominated social policy throughout the last three decades and disquiet over the Eleven Plus system was somewhat muted and low key. Moreover, the British Conservative Government were keen to retain the exam in Northern Ireland and the system of Direct Rule which existed from 1972 and lasted until a Labour Government was elected in 1997 facilitated the retention of the exam. Throughout this period there were a number of attempts to provoke debates about the benefits of the system for Northern Ireland children as a ploy to argue for the reintroduction of the grammar/secondary system in other parts of the United Kingdom. Much was made of the tendency for more Northern Ireland children to go on to university than any other region of the United Kingdom. Hence Northern Ireland students were upheld as a role model for United Kingdom students where comprehensive schooling could not match the high results of Northern Ireland’s grammar schools. However, the other side of the coin, the large numbers of Northern Ireland children (highest in United Kingdom) who left school with no qualifications was given minimal coverage. Hence, grammar schools were good for those children (mainly middle class) who managed to attend them while secondary school (mainly comprising working class) children fared poorly compared to their United Kingdom counterparts. The success of the labour party in the 1997 elections led to a renewed debate on the Eleven Plus system and the newly appointed Labour MP for Education instigated a review into the Eleven Plus system. The subsequent Gallagher report (Gallagher & Smith, 2000) was extremely critical of the exam suggesting that it scored poorly in its ability to predict which children were likely to benefit from a grammar school education. Under the Northern Ireland Assembly set up as part of the Good Friday Agreement,1 Martin McGuinness from Sinn Fein was appointed as Minister for Education in 1999. He set up a review body, the Burns Commission (Report by the Post-Primary Review Body, 2001) to consider the future of the Eleven Plus system and decide whether it should be retained or abolished and if abolished what would replace it. At the time of writing the UK government

158

MADELEINE LEONARD

have given a commitment to abolish the exam by 2008. Until then, the majority of children in Northern Ireland will continue to sit the test. It is against this background that the research reported in this paper was carried out. I noted with surprise that despite the existence of this test for over 50 years, children had never been asked to recount their experiences of what it was like to sit this test or asked their general attitudes to the subsequent division of children into “smart” and “stupid” that this test often entailed. Moreover, given that the test had moved centre stage to education policy in Northern Ireland, I was surprised at how little children were considered in the decision-making process regarding the future of the test. Hence I approached Save the Children, one of the primary children’s organisations in Northern Ireland to see if they would be interested in funding a piece of research into children’s experiences of sitting this test. The Department of Education for Northern Ireland was also approached and they agreed to co-fund the research. It was agreed that Save the Children would be responsible for the launch of the Report, which would take place at a popular public venue in Belfast. The organisation was able to secure the attendance of the Minister of Education and members of the Burns Commission at the launch and Martin McGuinness and the Chair of the Commission, Gerry Burns both agreed to make a public response to the Report at the launch. Hence, the research would feed the attitudes and experiences of children doing this test directly into the policy-making arena. One of the key aims of the research was to enable children’s voices to be heard in the debate on the future of the Eleven Plus. Hence my intention was to make children visible to the policymakers involved in the decision making process. I utilised Hart’s (1992) “ladder of participation” as a model for conducting research with children. After outlining the ways in which this was implemented, the remainder of the paper will consider some of the problems of involving children in policy related research in Northern Ireland.

APPLYING HART’S MODEL OF EFFECTIVE RESEARCH As I indicated earlier, I approached Save the Children and the Department of Education to fund the research on the basis that children were the experts on the Eleven Plus system and therefore their views and attitudes to it should form a fundamental part of the review process. Hence, children themselves were not involved at this stage of the research process. In this respect then, the research did not reach the seventh or eight’s rung of Hart’s ladder, which specifies that children themselves should initiate the research. Rather the research would be placed on Hart’s fourth to sixth rung of participation in that I tried to develop a research

Involving Children in Social Policy

159

design that enabled children’s voices to be heard by the relevant policy-makers. Collectively, the aim of these levels of participation is to move children from being the objects of research to being the subjects of research. However, a number of difficulties were encountered in implementing this model and the remainder of the paper is structured around highlighting these difficulties. Hart uses the term “Assigned but Informed” to define research whereby adults decide on the project and children volunteer to become involved or they know who decided that they should be involved. This throws up an initial problem in doing research with children. Recent policy in relation to children has a two-fold prong of protecting and promoting their interests. Because adults as parents or quasi-parents view their role as protectors of children from other unfamiliar adults, there are all sorts of limitations placed on adult researchers’ access to children. In particular, access to children has always to be negotiated through adults who act as gatekeepers on their behalf. The most common gatekeepers are parents and teachers and it is through gaining the consent of these adults that one gains the consent of children. Hence children’s consent is always secondary to adults’ consent. Indeed there is a danger in assuming that adult consent equals children’s consent. The research was fundamentally influenced by these limitations. Children were accessed through schools as this is where preparation for the Eleven Plus takes place and the intention was to follow children in eight primary schools through the process of preparing for the test, to sitting the test and waiting on the results. Access to children was gained through the class teacher and school principal in all the schools participating in the research. In negotiating this access, the usual guarantee of anonymity was offered. Hence the children who were to be the subjects of the research were invisible in the early stages of negotiating access. Teachers and then parents consented on children’s behalf by filling in permission letters sent out by the schools. While, we2 were very careful not to equate teachers’ consent with children’s consent, this creates an initial problem in that it may be difficult for children to play an active role in withholding their consent since influential adults have already consented on their behalf. As Denscombe and Aubrook (1992) point out, the relationship between adults and children within the school setting is profoundly unequal and children may experience hidden pressure to concede to what the teacher has already agreed to. Indeed, Morrow and Richards (1996, p. 95) suggest that it is not enough to seek the informed consent of children, the research has to go beyond this by creating opportunities for children to practice their “informed dissent.” Hence we went beyond actively seeking the permission of children to take part in the research by explaining to them carefully what the research was about, what we hoped to gain by doing the research and how we were going to feed the results back to them along with creating opportunities for them to withdraw their consent. We did this

160

MADELEINE LEONARD

by telling children that when they left the classroom to take part in focus group interviews, they could refuse to co-operate and just sit and listen to other children or read a book or engage in some other activity. In this way the children could maintain teachers’ approval for being helpful to visitors but simultaneously have the option of not taking part. In practice, no child decided to withdraw consent. We felt that we were practising Hart’s fifth rung on the “ladder of participation” whereby he suggests that children should be consulted about the research and have full understanding of the process and their opinions taken seriously. Moreover, we gave children the opportunity to withdraw their consent from engaging in the research. After gaining access we had to decide on what methods to use to obtain children’s views on the Eleven Plus. Interaction between adults and children is often influenced by the wider power relationships that exist between both groups throughout society. Morrow and Richards (1996) argue that employing a variety of techniques that allow children to feel a part of the research process can reduce the power of adults. Here we were influenced by Hart’s sixth rung whereby he suggests that children should be involved in the planning and implementation of the research. We tried to do this by using a variety of different methods for obtaining children’s views on the Eleven Plus and by allowing them to identify the issues rather than them being predefined by us. We used three specific methods: focus group interviews, stories and drawings. I want to elaborate briefly on each of these methods. We decided to use focus group interviews as some researchers suggest that oneto-one interviews may be too pressuring for children (Mauthner, 1997). This is because children often experience being interviewed by adults when they have done something wrong or when they perceive right answers are expected from them. Moreover, the unequal power relationships that exist between adults and children may be very difficult to overcome during one-to-one interactions. Hence, we decided to interview children in focus groups. These are regarded as giving children more power to set the terms of discussion and talk about their own views (Hurley, 1998). We spoke to children in focus groups on three occasions. On each occasion children were interviewed in the exact same focus group comprising four to six children. Other research suggests that between five and six is the optimum group size for group discussions with children (Greenbaum, 1987; Hill, Laybourn & Borland, 1996). Moreover, as in Hill, Laybourn and Borland’s research (1996), all the children in the focus groups came from the same class and knew each other outside the focus group. This may have facilitated discussion, which may have been more problematic if the members of the focus groups had been strangers. According to Morgan et al. (2002, p. 17) “groups of friends may feel empowered and supported in the co-presence of those they know.”

Involving Children in Social Policy

161

We asked children if we could record the discussions and assured them that we would be the only ones that would listen to the tapes. We also used a variety of techniques to make the focus group interviews less intimidating by playing with the tape recorder, allowing children to hear their voices on the tape, being on first name terms with the children and sitting on the floor with them during the interviews. Our rationale here was to try and dilute the hierarchical adult-child relationship (Morgan et al., 2002). Rather than ask direct questions we asked children to recall events for us such as “the day they sat the test,” “the night before the results were published” and how they felt when they opened their letters informing them of how they had performed in the test. Other research suggests that asking children to recall specific events is an effective approach to capturing children’s perspectives on their everyday lives (Mauthner, 1997). Brannen (1992) argues that given children’s relative powerlessness in society, it may be beneficial to utilise a range of different methodologies rather than rely on one approach. Hence, in addition to focus group interviews, we invited children to write stories and draw pictures on their thoughts on the Eleven Plus. We decided to use these additional methods because as childhood researchers have pointed out; children are often very familiar with communicating their thoughts and attitudes through these mediums (Hill, Laybourn & Borland, 1996). Our intention here was not to implement child-friendly methods on the assumption that children are a different species from adults, but to recognise that children may have different (not lesser) competencies compared to adults. Therefore it makes sense to draw on skills that they commonly practice in their daily lives (Christensen & James, 2000; Punch, 2002). We asked children to write two stories on their experiences of doing the Eleven Plus. The first story was written early on in the research and we used this story to structure the themes of the focus group interviews. Hence, we feel that these themes were identified by the children rather than by the researchers. We also asked the children to draw a picture for us on their “Thoughts on the Eleven Plus” and we decided to turn this into a competition with the winning picture going on the cover of the Report. This was to give children further ownership of the Report. We used these two methods because children are often used to communicating through drawings and stories rather than taking part in interviews. The research revealed a range of mainly negative experiences of children doing the test. For example, out of 193 pictures on the general theme of “Doing the Eleven Plus,” one in five children portrayed themselves crying. In their stories relating to sitting the actual exam, many children spoke of feeling nervous doing the test. One child described her experiences as follows: The night before the first exam I was so nervous and my hands were shaking and my legs were like jelly. I was very nervous and I did not want to do the test. That was the most sleepless night

162

MADELEINE LEONARD

I have ever had. As I was getting dressed on the morning of the exam, I was thinking of the dream I had that night. I got dressed and walked downstairs. Suddenly the phone rang. It was my uncle to wish me good luck. I was still as nervous as ever. As I was walking to school I started to think what the exam would be like, whether it would be hard or easy. I finally got into school, still shaking. I sat down in my seat. I couldn’t believe it was finally here, the Eleven Plus. Everyone was shouting and yelling and saying good luck to each other, while I sat so nervous, I couldn’t seem to speak.

According to Morrow and Richards (1996) if research is to be regarded as ethical, it is necessary where possible to report back to respondents. We tried to do this informally and formally. From the outset, we explained the purpose of the research and how it would be written up and disseminated. At each stage of the research, we reported back to the children on our understanding of the previous focus group interview and the themes we had identified in their stories and drawings. We told the children that Save the Children, at a major venue in Belfast, would launch the Report and that all the children who took part in the research would be invited to the launch. As mentioned earlier, we had secured the agreement of the Minister of Education and the Chair of the Review Body to attend the launch. We also told the children that after hearing their views as expressed in the Report, the Minister and Chair of the Review Body would respond. During the launch, a local children’s drama group also presented a short piece on the findings of the research using direct quotes from the focus group interviews with the children. We felt that this feedback was vitally important and we wanted children to see and be involved in the various stages of the research process. The main aim of the research was to try to ensure that the children’s voices would be listened, to by the relevant adults who would make decisions concerning the children’s lives. Hence we feel that the study to some extent provides a beneficial example of doing research with children. We encountered a further two problems which illustrate the extent to which adults control the extent to which children are able to practice agency in the research process.

ADULTS’ INVOLVEMENT IN RESEARCH WITH CHILDREN The Eleven Plus system is one of the greatest reproducers of class differences in Northern Ireland and there are a number of vested interests in keeping separate educational systems for working class and middle class children. This meant that although the schools that took part in the research were guaranteed anonymity, some of the school principals were very concerned that the Report might be critical of the Eleven Plus. This emerged in our initial interviews with principals where

Involving Children in Social Policy

163

we were often questioned about our own views on the Eleven Plus. We were able to convince all the principals that our purpose was to reflect children’s views on the exam and not our own. In some cases, this was difficult to accomplish, as there appeared to be general distrust that research would reflect the views of adult researchers rather than child subjects. The Report was subsequently critical of the exam as the majority of children we interviewed found the experience of sitting the test very stressful. However, it became apparent that some of the principals who agreed to their school participating in the research assumed that children would have a more positive attitude to the exam. This became an issue in relation to one school where the principal was very surprised to find that children from the school were critical of the test. The school in question takes great pride on the number of pupils who pass the Eleven Plus and has strong links with the local grammar school that these children subsequently attend. The principal became very agitated over the school being potentially identified as not supporting the retention of the Eleven Plus exam and wanted to disassociate the school from the research findings even though the majority of children from the school expressed concern about the test. This became problematic when it emerged that the picture selected to go on the cover of the Report was drawn by one of the pupils from the school. Since 193 children had drawn pictures for us, we felt that the pupil in question would like to have her name put on the inside cover of the report to denote her achievement. While we had guaranteed anonymity, the girl did not have an unusual name and it would have been difficult to associate the name of the school taking part in the research from the girl’s name. We approached the principal to gain the pupil’s consent along with her parents for her name to be placed on the inside cover of the report. The principal refused permission on the girl’s behalf without involving the girl or her parents in his decision. We tried to no avail to encourage the principal to allow the girl herself to make the decision but in this instance the adult gatekeeper rendered powerless one of the subjects of our research. In the end, we arranged for the pupil to receive a personal copy of the report with her name on the front cover. We also gave the pupil a framed copy of the cover of the Report with her name clearly visible. However, at the end of the day, we were unable to challenge the authority of the adult gatekeeper and his location as the primary decision-maker. The second problem that emerged concerned the Minister for Education, Martin McGuinness who was a member of Sinn Fein. As Minister for Education, Martin McGuinness would make the final decision on whether the Eleven Plus should be abolished or retained. Hence, we viewed his agreement to attend the launch of the Report as an effective way of enabling children’s views to feed into policy. However, because of his membership of Sinn Fein, the principal of one of the Protestant schools who took part in the research indicated to us that he would not allow the children to attend the launch of the Report, as he would not sit in the

164

MADELEINE LEONARD

same room as “that man.” We had already told the children of our plans for the launch and none had expressed any concerns about the attendance of the Minister. Indeed, during our final focus group interviews, we encouraged the children to engage in role play where they had to pretend that they had to present their views to the Minister of Education and Chair of the Review Body on the Eleven Plus. We referred to both of these people by their full names and the children from this particular school either did not recognise the name of the Minister for Education, did not associate him with Sinn Fein or were not bothered if they did. As in the previous example, the adult gatekeeper took this decision without consulting the children concerned and their ability to practice agency was rendered irrelevant. On the day of the launch, children attended from the other seven schools that took part in the research while pupils in the eighth school, were excluded from dissemination on the basis of a decision made by the adult gatekeeper. These two examples remind us, that children’s agency is often comprised by adults who may override the ability of children to practice their agency. Harden et al. (2000) argue that children live their lives under adult surveillance and adult control. Hence, “there is no free and autonomous realm of childhood outside the social relations in which childhood in general, and particular individual childhoods are forged” (Harden et al., 2000, p. 1.3). Within school-based research, adult teachers place themselves as the ultimate decision-makers over children’s participation in research not just at the access stages but also throughout the research process. This may fundamentally damage the ability of researchers to enhance children’s participation and control of the research process.

CONCLUSION While Hart’s “ladder of participation” provides a useful model for encouraging the effective participation of children in the research process, he may have underestimated the extent to which adults in their role as gatekeepers may impact on the intentions of researchers to promote children’s successful participation. Hence, researchers who argue for the adoption of Hart’s model tend to focus on the power imbalance between child respondent and adult researcher and leave undeveloped the power relationship between adult researcher and adult gatekeeper (Boyden & Ennew, 1996). The issue of children’s agency throughout the research process remains under-researched. Most researchers see the issue in terms of ensuring that children’s consent is gained after initial adult consent has been sought (Corsaro & Molinari, 2000; O’Kane, 2000). However, access is only the first part of the research process. Adult gatekeepers can withhold information from children about the research process and their role within it. Hence, subsequent decisions after

Involving Children in Social Policy

165

initial access continue to be made by adults rather than by children. Of course adults may claim that they have the child’s best interests at heart and their actions stem from attempts to protect children rather than deny them their rights. But, what happens if adults feel they are the best judges of children’s interests? The paper drew on two examples where adult gatekeepers located themselves as the sole decision-makers in relation to the children in their charge. On both occasions, the researcher had to interact and negotiate with the adult gatekeeper while the subjects of the research were left uninformed and irrelevant to the decisionmaking processes that came into operation. Because, research with children continues to operate through bargaining with adult gatekeepers, the extent to which researchers can guarantee children effective participation remains problematic. Hence, researchers need to illuminate in their accounts how children are located in the wider society in relation to adults. Indeed, Mayall (2000) argues that the generational order based on inequality between adults and children is the most significant obstacle confronting adult researchers on children and childhood. For children to more effectively participate in research, adult gatekeepers and not just childhood researchers need to be willing to share power with children. Yet, as the paper demonstrates, many adults continue to be unwilling to do this and their concerns overshadow those of children. There are no easy solutions to these issues but they do need to be debated. Children’s participation and the ability of adults to render children powerless and voiceless at all stages of the research need further exploration.

NOTES 1. The Good Friday Agreement (also known as the Belfast Agreement) provided for a 108-member Northern Ireland Assembly elected by proportional representation. The agreement was endorsed by 71% of voters in Northern Ireland and 94% of voters in the South of Ireland in May 1998 and the institutions were formally established in December 1999. The Assembly was suspended in 2003 and at the time of writing, Northern Ireland continues to be directly ruled by Westminster. 2. A research assistant was employed for four months.

ACKNOWLEDGEMENTS I would like to thank Save the Children and the Department of Education for Northern Ireland for funding the research. I would like to acknowledge the assistance of Ciara Davey who was employed as a research assistant for four months on this project.

166

MADELEINE LEONARD

REFERENCES Alderson, P. (1995). Listening to children: Children, ethics and social research. Buckingham: Open University Press. Alderson, P. (2001). Research by children. International Journal of Social Research Methodology, 4(2), 139–153. Arnstein, S. (1969). Eight rungs on the ladder of citizens’ participation. Journal of American Institute of Planners. Barn, G., & Franklin, A. (1996). Article 12 – Issues in developing children’s participation rights. In: E. Verhellen (Ed.), Monitoring Children’s Rights. The Hague: Martinus Nijhoff. Boyden, J., & Ennew, J. (1996). Children in focus: A manual for participatory research with children. Sweden: Save the Children. Brannen, J. (1992). Mixing methods: Combining qualitative and quantitative approaches. London: Gower. Brannen, J., & O’Brien, M. (1995). Children in families: Research and policy. London: Falmer Press. Christensen, P., & James, A. (2000). Research with children: Perspectives and practices. London: Falmer Press. Corsaro, W. A., & Molinari, L. (2000). Entering and observing in children’s worlds: A reflection on a longitudinal ethnography of early education in Italy. In: P. Christensen & A. James (Eds), Research with Children: Perspective and Practices. London: Falmer Press. Daniel, P., & Ivatts, J. (1998). Children and social policy. London: MacMillan. Denscombe, M., & Aubrook, L. (1992). It’s just another piece of schoolwork: The ethics of questionnaire research on pupils in schools. British Educational Research Journal, 18, 113–131. Gallagher, T., & Smith, A. (2000). The effects of the selective system of secondary education in Northern Ireland. Northern Ireland: Department of Education. General Assembly of the United Nations (1989). United Nations convention on the rights of the child. www.un.org. Gogle, F., & Tasker, G. (1982). Children and housework. Family Relations, 31, 395–399. Goodnow, J., & Delaney, S. (1989). Children’s household work: Task differences, styles of assignment, and links to family relationships. Journal of Applied Developmental Psychology, 10, 209–226. Greenbaum, T. (1987). The practical handbook and guide to focus group research. Lexington, MA: Lexington Books. Harden, J., Scott, S., Backett-Millburn, K., & Jackson, S. (2000). Talk, won’t talk? Methodological issues in researching children. Sociological Research Online, 5(2), 1–16. Hart, R. (1992). Children’s participation from tokenism to citizenship. Florence: UNICEF. Hill, M., Laybourn, A., & Borland, M. (1996). Engaging with primary-age children about their emotions and well-being: Methodological considerations. Children and Society, 10, 129–144. Hill, M., & Tisdall, M. (1997). Children and society. London: Prentice-Hall. Hurley, N. (1998). Straight talk: Working with children and young people in groups. York: Joseph Rowntree Foundation. Landsdown, G. (1995). Taking part: Children’s participation in decision-making. London: Institute of Public Policy Research. Lansdown, G. (1994). Children’s rights. In: B. Mayall (Ed.), Children’s Childhoods: Observed and Experienced. London: Falmer Press. Leonard, M., & Davey, C. (2001). Thoughts on the eleven plus. Belfast: Save the Children. Mauthner, M. (1997). Methodological aspects of collecting data from children: Lessons from three research projects. Children and Society, 11, 16–28.

Involving Children in Social Policy

167

Mayall, B. (2000). Conversations with children: Working with generational issues. In: P. Christensen & A. James (Eds), Research with Children: Perspective and Practices. London: Falmer Press. Milner, P., & Carolin, B. (1999). Time to listen to children. London: Routledge. Morgan, M., Gibbs, S., Maxwell, K., & Britten, N. (2002). Hearing children’s voices: Methodological issues in conducting focus groups with children aged 7–11 years. Qualitative Research, 21, 5–20. Morrow, V., & Richards, M. (1996). The ethics of social research with children: An overview. Children and Society, 10, 90–105. O’Kane, C. (2000). The development of participatory techniques: Facilitating children’s views about decisions which affect them. In: P. Christensen & A. James (Eds), Research with Children: Perspective and Practices. London: Falmer Press. Pole, C., Mizen, P., & Bolton, A. (1999). Realising children’s agency in research: Partners and participants? International Journal of Social Research Methodology, 19, 39–54. Punch, S. (2002). Research with children. The same or different from research with adults? Childhood, 9, 321–341. Qvortrup, J. (1987). Introduction: The sociology of childhood. International Journal of Sociology, 17, 3–37. Report by the Post-Primary Review Body (2001). Education for the 21st century. Northern Ireland: Department of Education. Shier, H. (2001). Pathways to participation: Openings, opportunities and obligations. Children and Society, 15, 107–117. Smith, A., Birthistle, U., Montgomery, A., & Farrell, S. (2000). Teachers and selection in Northern Ireland. In: T. Gallagher & A. Smith (Eds), The Effects of the Selective System of Secondary Education in Northern Ireland. Northern Ireland: Department of Education. United Nations Committee on the Rights of the Child (1995). Concluding observations of the committee on the rights of the child: United Kingdom of Great Britain and Northern Ireland. Consideration of Reports Submitted by State Parties under Article 44 of the Convention. CRC/C/15/Add.34. Verhellen, E. (1997). Understanding children’s rights. Gent: University of Gent Children’s Rights Centre. White, L., & Brinkerhoff, D. (1981). Children’s work in the family: Its significance and meaning. Journal of Marriage and the Family, 43, 789–798.

THE GENERATIONING OF POWER: A COMPARISON OF CHILD-PARENT AND SIBLING RELATIONS IN SCOTLAND Samantha Punch INTRODUCTION Although there has been much psychological research about children’s sibling relations, it has been a neglected area of study in sociology (exceptions are Brannen et al., 2000; Kosonen, 1996; Mauthner, 2002). This paper, based on empirical research on siblings in Scotland, explores the nature of the generational power structure within families from children’s perspectives. Childhood is a relational concept which forms part of the generational order. Alanen explains this as “a complex set of social processes through which people become (are constructed as) ‘children’ while other people become (are constructed as) ‘adults’ ” (2001, pp. 20, 21). Generational processes shape the nature of child-parent relations (Mayall, 2002). Alanen states that: one position (such as the parental position) cannot exist without the other (child) position; also what parenting is – that is, action in the position of a parent – is dependent on its relation to the action “performed” in the child position, and a change in one part is tied to change in the other (Alanen, 2001, p. 19).

In other words, child-parent relations are based on the understanding that childhood is relational with parenthood (see also Mayall, 2002). Alanen (2001) argues that

Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 169–188 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10009-3

169

170

SAMANTHA PUNCH

the social construction of childhood and adulthood involves a process, including the agency of both children and adults, which she refers to as a set of “practices”: It is through such practices that the two generational categories of children and adults are recurrently produced and therefore they stand in relations of connection and interaction, of interdependence (Alanen, 2001, p. 21).

These practices of generationing may be “childing” practices through which people are constructed as children or “adulting” practices through which a distinct adult position is produced. The ways in which children in the present study talked about the differences between their relationships with their parents and their siblings indicated that there are a range of generationing practices that take place within families. They referred to particular kinds of behaviour that were acceptable to engage in with other children (in this case with their siblings) but not with their parents. Overwhelmingly the key issue which children highlighted as distinct between their relations with parents and siblings was the differential nature of power in these relationships. Whilst it is not surprising that children perceive the distribution of power to be more unequal between children and parents than between siblings, the aim of this paper is to explore the nature of this power and how it is experienced from children’s point of view. In particular the paper discusses the ways in which children perceive child-parent relations compared with their sibling relationships in relation to the giving and receiving of power within the home. I recognise that by making a broad comparison between these two kinds of relationships, there is a danger of homogenising them and overlooking diversity. As Dunn points out “child-parent and child-sibling relationships are both complex and multidimensional” (1993, p. 90). Children’s experiences of sibling relationships vary according to their age, gender and birth order (Punch, 2001), and gender also shapes adults’ experiences of being a parent (Morgan, 1996). However, these intersections of birth order, gender and age are not going to be examined in this paper, but they are discussed elsewhere (Punch & McIntosh, 2003). Since the children spoke quite distinctly about the ways in which power is exercised between siblings compared with child-parent interactions, this paper focuses on a broad comparison of these two sets of relationships whilst acknowledging the complexity and variety within them. As the paper concentrates on an exploration of power relations within families, much of the discussion dwells on issues of conflict and control. Whilst the home can be a battleground where power struggles are played out, it is worth remembering that this is just one aspect of family relations and families can also include relationships of love, affection, caring and support (Brannen et al., 2000; Morgan, 1996). The paper begins by outlining the methods used and the definitions of power. It then discusses parental power in relation to legitimacy, household resources and

The Generationing of Power

171

children’s anticipated reactions of adult discipline. The nature of sibling power is highlighted before exploring the reciprocal expectations of sibling and childparent interactions. The paper ends by suggesting that the generationing of power relations can lead to differing degrees of backstage and frontstage performances within the home (see Goffman, 1959).

METHODS The study began with an exploratory phase of essay-based classroom research at three local Scottish schools where 180 children (aged 7–14) wrote essays about their experiences of sibling relationships. This stage informed the design of semistructured interviews and, along with some additional snowballing, enabled access to be negotiated with a sample of 30 families with three children between the ages of 5 and 17. In these families there are not particularly large age gaps between siblings, thus it is acknowledged that in other families with wider age differences, some of the generational power issues discussed in this paper may become more relevant between siblings as well as between children and parents. Each of the 90 siblings were interviewed individually in their homes, followed by 30 focus group interviews with all three siblings together. The group interview allowed for some sibling interaction to be observed and for group discussion of issues raised in the individual interviews. The group interviews followed the individual interviews so that any unequal power relationships between the siblings would not influence the research agenda for the individual interviews. The children were all full siblings of mixed socio-economic backgrounds, mostly living with both of their biological parents except for four single mother households. Consequently the limitations of the sample size mean that the impact of social class and different household forms cannot be fully explored. Nevertheless, the indepth and exploratory nature of the study has facilitated an examination of the kinds of processes involved in children’s understandings of sibling relations. All of the interviews included task-based methods such as spider diagrams, ranking exercises and the secret box technique (see Punch, 2002a) in an attempt to minimise unequal power relations between the adult researcher and child participants (see Punch, 2002b). In this paper, when using data from the interviews with children, in order to identify the quotations I use a pseudonym to protect the child’s identity, followed by their age, number of family interviewed and birth order (“o” indicates oldest, “m” is middle, and “y” is youngest child). To differentiate between the individual and group interviews, I add “Gp” at the end to indicate if the quotation was said during the group interview. In order to avoid confusion

172

SAMANTHA PUNCH

between parents and siblings, when it is unclear in the quotations I use a word in square brackets to emphasise the relationship of the person mentioned.

GENERATIONAL POWER PROCESSES WITHIN THE HOME During the interviews I asked whether children’s relationship with their parents is similar or different to their relationship with their siblings. Nearly three quarters of the children considered it to be very different. Most of the remaining quarter felt that there were some similarities but also differences and only four children said that the relationships were similar. However, although four children said that their relations with parents were “not really different” to their sibling relationships, at other points in these interviews examples were given to indicate that the relationships were possibly different. For example, Nigel said his relationship with his siblings and parents were similar: I behave the same with both of them, but earlier he had suggested that his younger brother: thinks he can do what he wants with mum not in the house (Nigel 9, Fam 21m), thereby implying that as an older sibling, and unlike his mother, he lacks legitimate authority. Similarly, the other three children who said that their relations with parents and siblings were similar, also discussed instances of parental power which appeared to be different from sibling power: my mum comes up and gives whoever’s been horrible a row, and then my brother will stop (Edward 9, Fam16m). Hence although in response to a direct question four children felt that their relationships with parents and siblings were similar, there was also evidence that even in these interviews some differences were implicit. Thus most of the children explicitly said that they treated their parents and siblings quite differently. Overwhelmingly the main contrast which they distinguished between their relations with parents and siblings was the differential nature of power in these relationships. Children perceived the imbalance of power to be strongly marked in inter-generational child-parent relations (see also Hartup, 1992) but lessened somewhat in the intra-generational sibling relationship (see Furman & Buhrmester, 1985). Power can be generally understood as “getting what you want” (Dowding, 1996, p. 50), in particular “power is the capacity to influence others” (Scott, 2001, p. 138). However, only a few children directly spoke using the terms “power” or “authority” (see also Thomas, 2000, p. 143): . . . well mum and dad can do things, well if Tom, all Tom can do is like go upstairs, tell on us and then it’s mum and dad who have more power. So it’s not like Tom has the power, it’s mum and dad that have the power so like you wouldn’t go annoying them (Robin 9, Fam13mGp). Mum’s got more authority (Rosemary 14, Fam9m).

The Generationing of Power

173

Most of the children expressed notions of power by talking about a person’s capacity to enforce their will: Well, I think your mum and dad can force you to do things and if you’re brothers and sisters you can’t force them to do things (Rhoda 12, Fam27o). Their indirect discussions of power were in line with Westwood’s definition of individual power as “the will to effect changes in another actor’s behaviour, context or view of the world” (2002, p. 14). Albrow argues that “power is normally structured in social relationships in a particular way, namely that one or more persons accept commands from others” (Albrow, 1990, p. 167; see also Smart, 1985, p. 73). The children in this study provided much evidence that in child-parent relations, parents tend to exercise more power over children, whereas in sibship, power is less linear and more contested. However, power is multi-dimensional and ubiquitous (Lukes, 1974; Westwood, 2002) and as Foucault reminds us, “Where there is power, there is resistance” (1979, p. 95). Children not only have strategies for counteracting adult power over their lives, but they are also active agents with an ability to assert power over adults (Valentine, 1999) even though such power tends to be limited and constrained. When broadly comparing sibship with child-parent relations, it becomes clear that adults’ generational location enables them to wield greater power (see also Buhrmester, 1992). As mentioned, whilst perhaps it is not a major revelation that parents have more power over children than siblings, it is important to consider how children understand these power differences. There are different forms of power and it can operate at both institutional and individual levels (Westwood, 2002, p. 14). This study is interested in interpersonal power where individuals significantly influence each other (Scott, 2001). Weber argued that this is the form of power characterised in parent-child relations (1968, p. 943), rooted in everyday face-to-face contexts of interaction. Thus the family is a context in which interpersonal power is exercised both between and within generations. The family is a social institution shaped by structural generational relations, based on a social and cultural construction of both childhood and adulthood. Recent developments in the sociology of childhood have questioned traditional models of socialization where adults are assumed to be in positions of authority over children (Hockey & James, 2003, p. 16). Valentine (1999) argues that children actively challenge parental authority and that nowadays families are more likely to be sites of negotiation rather than control and regulation. Nonetheless, as Brannen et al. explain, the generational distribution of power within families tends to remain unequal: In reconceptualising children as social actors, this is not to argue that children are now seen to wield more power vis-`a-vis adults, but rather to understand them as having the potential and the competences to exercise power. . . . Children’s lives are lived within the structural context of power in which adults regulate children’s bodies and minds. In their general status as children, and in their particular statuses as sons and daughters,

174

SAMANTHA PUNCH

children’s ability to act autonomously and their access to resources are constrained (2000, p. 178).

Children’s accounts of their family relationships indicate that families are structured by a generationing of power in which individuals draw upon resources attached to their social position: parents have more authority over me because, I’m not sure why, I think it might be because they’re adults (Lucas 12, Fam17m). The rest of this paper explores the nature of the generational power structure within families from children’s persectives.

PARENTAL POWER All of the children in this research agreed that parents are able to exercise more power over them than siblings. Their accounts of parental power indicate that parents have legitimate authority and control over resources, both of which are interrelated mechanisms that can be used as a form of disciplinary power. This section explores these interconnected aspects of the generationing of power, which are also linked to the ways in which children anticipate, and cooperate with, parental power. Legitimate Authority Dowding (1996) points out that legitimate authority is an obvious source of power since the force of law results in the likelihood of compliance by others. As Scott argues: A parent exercises interpersonal power over a child, but also has certain legal rights that the child may grow up to accept and that will be recognised by others (Scott, 2001, p. 30).

Parental power is linked to their functional roles as protectors and providers (Beetham, 1991, p. 45; Roffey et al., 1994, p. 7). In her research on childhoods in London, Mayall found that children perceived that: Parents rightly had authority over their children. Many young people also provided justification for parental authority – parents knew more than children and had a duty to protect them and provide for them; so it was for them to decide how life should be lived, including how children should behave (Mayall, 2002, p. 46).

In other words, there are positive and productive features of power relations (see also Hindess, 1996; Smart, 1985) as parents are expected to fulfil their responsibilities as carers. This sibling study also found that children tended to accept parental authority. When their parents asked them to do something

The Generationing of Power

175

they were much more likely to comply than if their siblings asked them. Erica explains that she is not very successful at telling her younger sister what to do: It doesn’t really work ’cos we haven’t got that much of an age difference, ’cos if I boss her around then she takes it as like me having a joke and she’s just going “oh shut up Erica”. But if it was mum or dad who said that she would like stop doing what they were giving her a row for or something. She doesn’t really take it as anything if I give her a row [a telling off], she doesn’t really (Erica 10, Fam15m).

Many of the children suggested that their obedience in response to parents or siblings was linked to a sense of what is right and wrong in the acts that are requested. On the whole parental requests were judged to be more acceptable compared with the more unreasonable demands of siblings: Like [sisters] tell me to go and get their socks and shoes and stuff and I just go “no you can go and get them yourself”. Mum and dad tell me to clean my teeth and stuff, have breakfast (Julianne 10, Fam7o).

Similarly Kathryn explains that her brother, Tony, and her father do not always grant her and her sister access to household resources and she interprets the legitimacy of their actions in different ways: Like if I wanted to phone mum for any reason he wouldn’t let me ’cos he’s on the Internet and we’ve only got the one line and if I picked up the phone and waited for a while it disconnects it and he like shouts at us . . . . Dad gives us a row for different things, not for wanting to phone mum or things like, ’cos Erica’s always on dad’s computer and she never asks, and sometimes it crashes. Like it goes dead and you can’t move anything and he has to restart it and stuff ’cos that, and he just like gives her a small row. And Tony’s shouting really loud and stuff (Kathryn 9, Fam15y).

These quotations indicate that childrens’ motivations of whether to abide by or resist commands is associated with whether or not they believe the individual has a right to make such a request. Parents are perceived to have legitimate power that is tied to their underlying parental responsibility for their children’s well-being (Beetham, 1991). As Scott argues, “Power is legitimate because it is accepted as being right, correct, justified, or valid in some way” (Scott, 2001, p. 20; see also Albrow, 1990, p. 164; Gerth & Wright Mills, 1991, p. 79). Many children remarked that they are more willing to adhere to parental rules or requests rather than those given by siblings: Interviewer: If your parents tell you to can you say no? Simon: No we usually just do it ’cos otherwise they just end up shouting at us and there’s no point (Simon 14, Fam19oGp).

176

SAMANTHA PUNCH

Because they think that our mum and dad are the boss and because they’re oldest and they’re adults, they should be in charge. But when my mum goes in the garden and my dad’s at work, I’ve got to be in charge because I’m normally told to be but they don’t like listening to me. So when I tell them to get out of the bath or something, they don’t do it, they only do it when an adult says so (Helen 10, Fam25o).

Thus, on the whole, children believe that parents have the right to tell them to do certain things, whereas siblings do not possess the same kind of legitimacy: Because he’s not an adult so he doesn’t have the right (Kathryn 9, Fam15yGp). Parental power is justified because of their generational and social position within the family.

Resource Power Power is based upon resources (Dowding, 1996; Martin, 1977) and interpersonal power operates through “the various resources on which some depend and to which others can give access” (Scott, 2001, p. 28). Within families, parents tend to control access to household resources, including income and material goods (O’Brien, 1995), as well as controlling much of children’s use of time and space (Ennew, 1994). Children also depend on their parents for “access to friends and to spaces and times outside the home” (Mayall, 2002, p. 48). Nearly all the children in this study confirmed that parents have greater control of household resources than siblings and they use these varied resources to encourage children to do as they ask: She [mum] could say something like “oh well I’m not taking you to athletics tonight” and then, but Douglas [brother] couldn’t say that (Rosemary 14, Fam9m).

The children are aware that parents have the capacity to manipulate the use of material resources within the home. Michael illustrates that if he disobeys his parents he would probably get a big big big ban from using my TV or the playstation (Michael 11, Fam29o). In particular, as Dowding reminds us: “Money is an obvious source of power” (Dowding, 1996, p. 65) and within households parents hold most economic power: . . . mum and dad pay for my martial arts so then if I usually get in trouble that’s the threat that they will stop paying for all that stuff (Simon 14, Fam19o). Mum and dad control the money flow (Tony 15, Fam15oGp).

Consequently parents’ legitimate authority is enhanced “through the use of resources that can serve as punitive and remunerative sanctions” (Scott, 2001, p. 13; see also Galbraith, 1983, p. 48). The children described a range of punishments

The Generationing of Power

177

that they know their parents may use, including being grounded, being sent to their room or to bed early, withdrawal of privileges such as use of computers or watching television, and decreased pocket money. In contrast, siblings do not have the same resources at their disposal: Robin: Well Tom couldn’t say “oh go to bed, go to your bed” like but mum and dad can. Susan: “You can’t have any tea”, they can do that. Interviewer: But can Susan never do that? Robin: No well she could tell us off but not in the way that mum and dad could, like we probably wouldn’t listen (Susan 12, Robin 9, Tom 5, Fam13Gp). Tim: My brother can’t exactly say “your bed time is 7 o’clock.” George: “You’re getting no pocket money.” Tim: Yeah, but your mum and dad can (George 11, Tim 9, Andy 5, Fam23Gp).

Therefore, children recognise that parents have the ability to act and enforce punishments. They believe that parents have a right to discipline them to some extent, and it is this belief and acceptance of their power that encourages children to obey their parents more than their siblings. However, as Albrow’s quote reminds us, children not only adhere to parents’ wishes as they respect their authority but also because they fear the consequences of not doing so: While belief in legitimacy was a major factor in enhancing the stability of a social order, for a great deal of the time individuals were oriented towards it in terms of expediency, either from fear for the consequences if they departed from it, or from the advantages they perceived if they conformed. Equally for much of the time people adhered to an order simply because they were accustomed to do so and could see no good reason for doing anything else (Albrow, 1990, p. 163).

In addition, as Albrow suggests, responding positively to a parental request may be partly due to an acceptance of routinised everyday interactions: Lucas: They [parents] can boss me around more. Interviewer: How do you mean? Lucas: Well if I have to clear a mess or tidy my room or do the washing-up then I have to do it. Interviewer: Can you not say no or . . .? Lucas: Well I do say no most of the time but I still have to do it . . . Interviewer: Is there any consequences say of not doing what they say? Lucas: No not really but I still do it anyway (Lucas 12, Fam17m).

Smart argues that “Discipline is a technique of power which provides procedures for training or for coercing bodies (individual and collective),” (1985, p. 85). Thus parents may assert their disciplinary power for several reasons, including training and socialising children, encouraging cooperation, and maintaining order and docility within the home (see also Brannen et al., 2000). This coincides with Foucault’s description of disciplinary punishment “that operates in the process of training and correction” (Foucault, 1977, p. 180). Parental power is linked

178

SAMANTHA PUNCH

to the parenting role: “it aims not only to constrain those over whom it is exercised, but also to enhance and make use of their capacities” (Hindess, 1996, p. 113).

Anticipated Reactions For disciplinary power to be effective, the threats of withdrawal of priveleges or enforcement of spatial or temporal sanctions must be perceived as credible (see Dowding, 1996). Children can be caused to act or prevented from acting in a prohibited manner by the knowledge that parents will carry out disciplinary action: Disciplined individuals have acquired skills, habits of action, desires, and qualities of character that allow them to act in appropriate and expected ways and to do so through the exercise of self-control (Scott, 2001, p. 94).

Many children in this study claimed that the anticipated reactions of their parents was enough to encourage them to behave in certain ways: We’re a lot more sensible when our mum and dad are there and that. Just in case we get into trouble or something (Simon 14, Fam19o). Well, I don’t argue as much with my mum and dad . . . . Because I know they can send me to my room and things like that (Rhoda 12, Fam27o).

Thus children’s expectation that parents will assert their authority and impose punishment can provide a motivation for certain forms of behaviour. In other words, discipline is a productive power that parents can exercise over children “to develop their capacity for self-control” (Hindess, 1996, p. 113; see also Brannen et al., 2000). In particular it can result in their cooperation rather than resistance. As Martin argues, “compliance may result from a fear of the consequences of non-compliance” (1977, p. 42). One of the reasons why parents are able to exercise more control over children than siblings is because power is rooted in relations of dependence (Emerson, 1962, p. 349). Child-parent relations are based on a dyadic interdependence where the child tends to be more dependent on the adult. Parents may depend on their children for emotional attachment (Hood-Williams, 1990) and for a contribution to the domestic division of labour (Morrow, 1996) but ultimately in the U.K. children are very dependent on their parent’s material resources (Mayall, 2002). In contrast, siblings experience a more mutual dependency on each other compared with the more unequal balance of power in child-parent relations. This paper is concerned with the broad comparison of the nature of power in child-parent relations and sibships, but obviously there is diversity within these

The Generationing of Power

179

different types of relationships. Furthermore, generational relations are cross-cut by gender, and power relations may differ according to the personal qualities of particular individuals. There is not scope in this paper to discuss these in detail, only to recognise that some variation exists in the power asserted by mothers and fathers. For example, as Mayall found in her study of London childhoods (2002, p. 45), fathers were sometimes described as the harsher authority figure: They’d [parents] probably be like the bosses, dad especially, dad’s more of a boss than mum. ’Cos if mum gives you a row you don’t really take it as much you just think “oh it’s only mum”, if dad gives you a row you’re like “OK” (Erica 10, Fam15m).

Furthermore, there are different parenting styles across families where material rewards and punishments are used to varying degrees. Similarly, children assert their own agency choosing to resist or comply with adult power to a varied extent. They reported a range of strategies and tactics for negotiating with both parents and siblings. This is consistent with Mayall’s (2002) findings that within the broad framework of parental authority; children can resist and assert control over their own use of time and space. Thus, generational identities are not rigidly fixed and they involve individual agency. Children are active agents in the construction of their childhood identities even though they do not necessarily experience equal capacity to assert power within the family. Nevertheless, despite recognising that children by no means passively accept parental power over their lives, broadly speaking when comparing child-parent relations with children’s sibling relations, a marked difference emerges in the nature and exercise of power.

SIBLING POWER Siblings do not possess legitimate authority, nor do they control household resources to the same degree as parents. Consequently, as children are not willing to readily accept their siblings as having power over them, “then authority breaks down into a cruder form of power, of threats and offers” (Dowding, 1996, p. 64). Sometimes older siblings are left in charge of their younger siblings when their parents are temporarily absent. However, despite being granted some legitimate power on behalf of the parents, the other siblings often do not recognise their authority: if I’m in charge of them they don’t actually do anything I say even if it’s for their own good (Tony 15, Fam15oGp). In this research, siblings are part of the same generation and are relatively close in age, resulting in them being less likely to take seriously a command or an admonishment from their brother or sister. Thus parental power cannot readily be conferred from parents to older siblings because the similarity of the generational location of siblings weakens the legitimacy of even a temporary transference of power during parental absence.

180

SAMANTHA PUNCH

Consequently different patterns of power take place between siblings. There tend to be greater power struggles and increased resistance that is more openly displayed. Power relations between siblings exist but operate in a different manner to parent-child relations. As siblings rarely have legitimate power that is recognised and accepted as right, they are more likely to use bargaining and physical power. Dowding argues that bargaining can take many forms and that, “Each side may try to make offers or threats with whatever resources they have available” (Dowding, 1996, p. 66). For children these resources may include information (for example, agreeing to keep secrets from parents), a loan of material goods such as toys, computer games or CDs, and labour such as offering to undertake household chores on their behalf: Well, normally I just ask her to do it for me and say I’ll do something else for her later (Nick 11, Fam20m). I’d just ask them loads of times until they said yes. After that, I’d kind of make a bargain with them, like a few sweets if they did it or something (Michael 11, Fam29o). Craig: Like you could say “If you tell mum and dad that I did that, then I’ll tell that you did this.” Interviewer: What sort of thing? Craig: Say I broke a glass, say I smashed a CD that dad had bought that was old, and then I don’t know, they hit somebody over the head at school and knocked them out. I could say “if you tell mum and dad that I smashed that CD then I’ll tell them that you hit someone at school” (Craig 11, Fam26m).

Siblings engage in more complex processes of negotiation and bargaining because it is harder for them to exercise disciplinary power compared with their parents. This is perhaps why sibling power struggles are also more likely to involve physical force. The children in this research recognise that parents have greater physical strength but also that it would not be considered appropriate to initiate a fight with them (see also Foot et al., 1980, p. 270; Hendry et al., 1993, p. 116): George: Well, you don’t usually fall out so much with your mum and dad that much . . . . Well, you don’t fall out the same because if you fall out with your brother you just start fighting them. Tim: You couldn’t just punch your mum and pull her hair! George: Because maybe if you did do it with your mum and dad, they’re much stronger and I don’t think they would punch you back. Tim: You’d get in trouble (George 11, Tim 9, Andy 5, Fam23Gp). And we hardly ever like fight with them, not like punch, punch, ’cos that would be, you never do that with your parents ’cos they’d like win (Erica 10, Fam15m).

The children described a range of physical fighting between siblings that takes place to varying degrees of intensity. According to their accounts, physical contact during

The Generationing of Power

181

verbal arguments play a comparatively prominent role in sibling disputes compared with the more verbal disagreements of child-parent relations. Thus power struggles between siblings include the use of physical force to a much greater extent than between children and parents. The nature of physical fighting depends on the age and gender of siblings but, as mentioned, a detailed discussion of the intersections of gender and birth order within individual sibships remains outside the scope of this paper. The key difference between children’s accounts of sibling power and parental power appears to fall in line with the two conceptions of power that Hindess identifies: power as a generalized capacity to act and power as “involving not only a capacity but also a right to act, with both capacity and right being seen to rest on the consent of those over whom the power is exercised” (1996, p. 1). In other words, the power which siblings exercise is “power as simple capacity” whereas parents operate with “power as legitimate capacity” (see Hindess, 1996).

ROUTINISED AND RECIPROCAL EXPECTATIONS The differential nature of power in sibships and child-parent relationships shapes the particular forms of interaction that take place between them. Goffman describes the interaction order as “the mutual self-presentations of embodied individuals as they construct and reconstruct their identities and life plans in response to each other. It consists of a complex of everyday encounters that may become more or less routinised” (Scott, 2001, p. 29; see also Goffman, 1959). Therefore, within families, children and adults develop expectations of each other’s behaviour during routinised social interactions. According to Scott, these encounters become “subject to constant marginal transformations of reciprocal expectations that are largely taken for granted by the participants” (2001, p. 29). The children’s accounts in this research illustrate that with parents they tend to experience mutually more positive treatment whereas with siblings their actions are more likely to be provocative and reciprocally negative. Many children remarked that they exerted more self-discipline with their parents and were more likely to behave better towards their parents than their siblings: I behave much better with my mum and dad . . . . Like I just behave much better, like I just sit down with my mum and dad, whereas with Richard and Ian I just like, they just come up and annoy, Richard comes up and annoys me and then I go and annoy Ian and then we all start in a fight (Angus 13, Fam11m). I’d probably be a little less mischievous (Nick 11, Fam20m). Well I act quite like nice and kind and things [with mother]. I wouldn’t like be really annoying in front of her (Douglas 9, Fam9y).

182

SAMANTHA PUNCH

The different generational positions lead children to behave with more self-restraint and respect towards their parents. Partly this is a fear of the consequences if they do not, and partly it is a reflection of their acceptance of parental authority. In addition, it is because the social relationship with their parents is based on mutually orientated positive action (see also Albrow, 1990): your mum and dad behave differently to you. They don’t do the same things (Graham 11, Fam4mGp). Younger siblings in particular highlighted that their parents were much more likely to treat them well: They’re [parents] all nice to me and there’s Edward [brother] sometimes he’s horrible to me and mummy and daddy have never been horrible to me (Julian 6, Fam16y). . . . like my mum and dad like I can say things to them and like they won’t, it sounds a bit weird but like they won’t tease me or anything. Like if I said it to my brother or sister they would probably tease me (Jason 10, Fam22y). It’s different ’cos they’re not horrible to me (parents) (Shaun 10, Fam30y).

Therefore, child-parent relations are based on a reciprocal expectation of more positive treatment which in turn leads them to get on better and argue less compared with sibships (see also Buhrmester, 1992). This is not to say that parents and children do not argue or treat each other badly at times, but comparatively speaking, it is less frequent than sibling quarrels: I don’t fight with my mum and dad, well sometimes but not all the time as much. Say I said to dad “can I borrow this” and he’ll say yes but the boys will say no (Roxanna 9, Fam26y). As this quote shows, parents also are perceived as potentially more cooperative than siblings. Many children said that child-parent relations are different from sibling relations because they do not fall out with their parents to the same extent or in the same way as with sibings: It’s quite different ’cos we don’t always fight with them . . . . ’Cos they don’t steal my stuff and hog the telly (Aleyda 12, Fam 5m). Nicola said that if she argues with her mother: I feel bad (Nicola 14, Fam20m) but that she does not feel the same when she fights with her siblings. Several children expressed similar sentiments: Rosemary: I don’t wind my mum up because I just wouldn’t . . . . Because she’d get angry or she’d, I don’t know . . . Interviewer: But wouldn’t Heather or Douglas if you wound them up? Rosemary: Yeah but I wouldn’t care but I care if my mum gets angry with me (Rosemary 14, Fam9m).

Children are more likely to regret arguing with parents than with siblings. There are several possible reasons for this: because they are more accustomed to conflict within sibling relations, there are greater negative consequences of falling out with parents and they have more respect for parents because of their generational and

The Generationing of Power

183

social position within the family. Most of the children agreed that they showed their parents greater respect than their siblings: I look up to them [parents] (Ian 14, Fam11o). . . . she [sister] just does these repeatedly annoying things. They [parents] just tell her to stop it but she’d listen to them more because she’s got more respect for them (Samuel 16, Fam20o). They don’t respect me ’cos if I told them off and said “you’re not allowed to do that”, they’d probably just, well Duncan especially, just keeps on doing it until I threaten to tell mum and then mum’ll be really angry (Michael 11, Fam29o). You give them more respect than your brother and sister, your brother and sister are like playmates (Erica 10, Fam15m).

As Goffman reminds us, “this show of respect may, of course, be motivated by a desire to impress the audience favourably, or avoid sanctions” (1959, p. 111). Consequently for all of the above mentioned reasons (respect, reciprocal positive treatment, parental authority and anticipated reactions), most of the children’s accounts indicated that they acted in a more restrained manner with their parents, more carefully managing their presentation of self (see Goffman, 1959). Some children suggested that this could be constraining as they felt more obliged to tailor their techniques of impression management. For example, Christian explained: Well you might be in a wee bit of a bad temper with your brothers, you show adults more respect . . . . ’Cos you know that your mum and dad could punish you if they wanted to and they could do more like that. Brothers they can’t. And they, you feel more free in front of your brothers and sisters than you do in front of your mum and dad (Christian 8, Fam29m).

As Christian’s quotation suggests, sibling interaction is similar to Goffman’s (1959) descriptions of backstage behaviour where a performer relaxes his/her personal front and may neglect social rules of politeness and decorum. Dunn describes the ways in which siblings love and hate, play and fight, tease and mock with a “devastating lack of inhibition” (1984, p. 11). There are fewer “rules” associated with sibling interaction, and conflict is almost expected as part of “doing” sibship (Brannen et al., p. 118; Buhrmester, 1992; Raffaelli, 1992). In contrast, children exert more self-control with parents, providing more of a frontstage performance where their behaviour tends to be more respectful and restrained. Thus, although sibling interactions tend to be more negative than those with parents, to a certain extent sibling encounters can be enabling, as they may allow children to more openly express their thoughts and feelings (Punch, 2002c). Furthermore, some of the children’s comments indicated that when they are with their siblings away from their parents, they are allowed to do things which their mothers or fathers would not permit (see also Foot et al., 1980, p. 270):

184

SAMANTHA PUNCH

Well sometimes when mum and dad aren’t around we do things that we’re not allowed to do in the house . . . . Maybe fiddle about with things (Michael 11, Fam29o). . . . they’ll [siblings] let you do stuff (Josephine 15, Fam5o). Interviewer: And what’s that like when Jessica’s in charge? Julian: Quite fun ’cos we don’t have mummy and daddy and we can do whatever we want for a wee while. Interviewer: What sort of stuff could you do that you might not do if your mum and dad were here? Julian: We wouldn’t have to tell mummy and daddy that I’m going outside and I can go on the scooter without mummy and daddy knowing (Julian 6, Fam16y).

On the one hand, backstage with siblings, children may be less inhibited and more able to openly express their feelings. On the other hand, because many social conventions are dropped, the relaxed atmosphere of the backstage can also lead to conflict as anger may be more easily vented (Punch, 2002c). When discussing sibling interactions, the two key words that reemerged consistently in the interviews were “annoying” and “fighting.” As mentioned, child-parent relations are characterised by comparatively less arguments than sibships. The children commented that one of the main differences between parents and siblings is that parents try to intervene and sort out arguments whereas siblings tend to cause them: we don’t really argue that much with mum and dad. They normally stop arguments (Nick 11, Fam20m). In other words, unlike parents, siblings are considered to be more irritating, often provoking each other into an argument on purpose: My mum and dad aren’t as annoying my three sisters (Julianne 10, Fam7o). I know I can be moody sometimes but I think they kind of encourage it. By just being irritating (Heather 16, Fam9o). Interviewer: Why do you fight less with your mum and dad? Roxanna: Well, mum doesn’t run around saying “Roxanna you’re an idiot!” Or dad. They really don’t annoy me as much (Roxanna 9, Fam26y).

However, the children did not only blame their siblings for being annoying, but they also recognised their own active role in initiating disputes. In the same way that children and parents engage in mutually more positive interactions, siblings behave in a reciprocally negative manner: because they’re really annoying to you so you are annoying back (Douglas 9, Fam9y).

CONCLUSION Therefore, we have seen that the nature of child-parent and sibling relations is shaped to a large extent by their generational position and the subsequent

The Generationing of Power

185

ways in which power is exercised and resisted in these relationships. Both intergenerational and intra-generational relationships within families involve power struggles and negotiations. Nevertheless, ultimately children are more likely to cooperate with or cede to parental power. As Valentine comments: “parents’ superior age, size and life experiences means that their power over their children is literally embodied” (1999, p. 150). Furthermore, children’s belief in parents’ legitimate authority can lead to self-disciplined and restrained behaviour as well as avoidance of certain forms of action because of the anticipated impact of parental reactions. In contrast, siblings cannot exert the same degree of disciplinary power. As a result, children do not feel the requirement to carefully manage their presentation of self during sibling interactions. Thus with siblings, on the one hand, the power they attempt to wield over each other is less effective and weaker than parental power: Well they could try to but it wouldn’t normally work (Nick 11, Fam20m). Yet, on the other hand, the “informality” of sibling interactions means that they do not have to strive to maintain a particular impression and are perhaps more free to express themselves with less fear of the consequences: Yeah because usually mum and dad it’s like we have to get on because if you want to do stuff and you want them to help you with stuff you’ve got to, whereas with brothers and sisters it’s a bit different . . . . It’s just a case of if you want to get on with them or not, if you want to be pals with them or if you want to just keep them out of your way, it’s a bit different (Josephine 15, Fam5o).

Goffman explores differences in backstage and frontstage performances by drawing mainly on examples from public arenas (such as shops or restaurants) rather than from private spheres. Similiarly, authors who have used his dramaturgical framework tend to focus on formal interaction in public places (for example Chung, 1990; Gillespie, 1987). This paper has shown how different degrees of performance can be enacted within informal relationships in the private domestic sphere which is often assumed to be a backstage arena. It has suggested that interaction with siblings consists of a backstage performance whereas with parents a greater degree of frontstage performance may be required. The child-parent relationship is informal and in comparison with more formal, public relationships such as with teachers or doctors, one could say that parents require less impression management. However, the value of comparing childparent relations with another informal relationship, in this case sibship, is that it highlights the ways in which children can more carefully manage their presentation of self with parents than with siblings. Interaction between children and parents and between siblings is shaped by their generational positions: what is perceived as acceptable behaviour between siblings is perhaps not appropriate with parents. Within the inter-generational child-parent relationship power is more likely to be exercised over children because of parents’ legitimate capacity which stems from

186

SAMANTHA PUNCH

their parental role as caregiver as well as their adult status. In contrast we have seen that within the intra-generational sibling relationship power tends to be more reciprocal and exerted as a cruder form of “simple capacity” (Hindess, 1996). The following final extract from one of the sibling group interviews sums up this generationing of power within family relationships: Roxanna: Your mum and dad take care of you. Craig: They’re not horrible to you. Interviewer: Why can you be horrible to brothers and sisters but not to parents? Craig: Because they’re like the same age as you. “You kissed each other last night, na na na na na na.” It’s stupid! Gareth: It’s okay with brothers and sisters. But with mum and dad, it’s just stupid. Craig: It’s like an adult, they’ll just think you’re stupid. Interviewer: But brothers and sisters won’t? Craig: You think they’re stupid but they’ll think you’re stupid too. (Gareth 13, Craig 11, Roxanna 9; Fam26Gp)

This quotation touches on the key themes of this paper and highlights the relational and reciprocal aspects of family relations. In order to understand the ways in which child-parent relations and sibships are shaped, we need to have an understanding of how power operates within these generationing processes. This paper has presented empirical data that illustrate the different ways in which power is played out in inter- and intra-generational relationships within families. Power is an integral part of these generationing processes that impact upon children’s experiences of family relations.

ACKNOWLEDGEMENT I offer grateful acknowledgement to the British Academy for their financial support of this study.

REFERENCES Alanen, L. (2001). Explorations is generational analysis. In: L. Alanen & B. Mayall (Eds), Conceptualising Child-Adult Relations. London: Routledge Falmer. Albrow, M. (1990). Max Weber’s construction of social theory. London: Macmillan. Beetham, D. (1991). The legitimation of power. Basingstoke: Macmillan. Brannen, J., Heptinstall, E., & Bhopal, K. (2000). Connecting children: Care and family life in later childhood. London: Routledge Falmer. Buhrmester, D. (1992). The developmental courses of sibling and peer relationships. In: F. Boer & J. Dunn (Eds), Children’s Sibling Relationships: Developmental and Clinical Issues (pp. 19–40). London: Lawrence Erlbaum Associates.

The Generationing of Power

187

Chung, Y. K. (1990). At the palace: Researching gender and ethnicity in a chinese restaurant. In: L. Stanley (Ed.), Feminist Praxis (pp. 189–204). London: Routledge. Dowding, K. (1996). Power. Buckingham: Open University Press. Dunn, J. (1984). Sisters and brothers. London: Fontana. Dunn, J. (1993). Young children’s close relationships: Beyond attachment. London: Sage. Emerson, R. (1962). Power-dependence relations. In: J. Scott (Ed.), Power (Vol. 2, pp. 343–361). London: Routledge. Ennew, J. (1994). Time for children or time for adults. In: J. Qvortrup, M. Bardy, G. Sgritta & H. Wintersberger (Eds), Childhood Matters: Social Theory, Practice and Politics (pp. 125–143). Aldershot: Avebury. Foot, H. C., Chapman, A. J., & Smith, J. R. (1980). Patterns of interaction in children’s friendships. In: H. C. Foot, A. J. Chapman & J. R. Smith (Eds), Friendship and Social Relations in Children (pp. 267–289). Chichester: Wiley. Foucault, M. (1977). Discipline and punish. London: Allen Lane. Foucault, M. (1979). The history of sexuality (Vol. 1). London: Allen Lane. Furman, W., & Buhrmester, D. (1985). Children’s perception of the personal relationships in their social networks. Developmental Psychology, 21(6), 1016–1024. Galbraith, J. (1983). The anatomy of power. London: Corgi Books. Gerth, H., & Wright Mills, C. (1991). From Max Weber: Essays in sociology. London: Routledge. Gillespie, J. (1987). The phenomenon of the public wife: An exercise in Goffman’s impression management. In: M. Deegan & M. Hill (Eds), Women and Symbolic Interaction (pp. 191–210). Boston: Allen and Unwin. Goffman, E. (1959). The presentation of self in everyday life. Harmondsworth: Penguin Books. Hartup, W. (1992). Friendships and their developmental significance. In: H. McGurk (Ed.), Childhood Social Development: Contemporary Perspectives (pp. 175–205). Hove: Lawrence Erlbaum Associates. Hendry, L., Shucksmith, J., Love, J., & Glendinning, A. (1993). Young people’s leisure and lifestyles. London: Routledge. Hindess, B. (1996). Discourses of power: From Hobbes to Foucault. Oxford: Blackwell. Hockey, J., & James, A. (2003). Social identities across the life course. Basingstoke: Palgrave Macmillan. Hood-Williams, J. (1990). Patriarchy for children: On the stability of power relations in children’s lives. In: L. Chisholm (Ed.), Childhood, Youth and Social Change: A Comparative Perspective (pp. 155–171). London: Falmer Press. Kosonen, M. (1996). Siblings as providers of support and care during middle childhood: Children’s perceptions. Children and Society, 10, 267–279. Lukes, S. (1974). Power: A radical view. Basingstoke: Palgrave. Martin, R. (1977). The sociology of power. London: Routledge & Kegan Paul. Mauthner, M. (2002). Sistering: Power and change in female relationships. Basingstoke: Palgrave Macmillan. Mayall, B. (2002). Towards a sociology for childhood. Buckingham: Open University Press. Morgan, D. (1996). Family connections. Cambridge: Polity Press. Morrow, V. (1996). Rethinking childhood dependency: Children’s contribution to the domestic economy. The Sociological Review, 44(1), 58–77. O’Brien, M. (1995). Allocation of resources in households: Children’s perspectives. Sociological Review, 43(3), 503–517.

188

SAMANTHA PUNCH

Punch, S. (2001). Household division of labour: Generation, gender, age, birth order and sibling composition. Work, Employment and Society, 15(4), 803–823. Punch, S. (2002a). Interviewing strategies with young people: The “secret box”, stimulus material and task-based activities. Children and Society, 16, 45–56. Punch, S. (2002b). Research with children: The same or different from research with adults? Childhood, 9(3), 321–341. Punch, S. (2002c). Children’s backstage and frontstage performances: A comparison of sibship and friendship. Paper presented at the XVth International Sociological Association World Congress, Brisbane, Australia, 7–13 July 2002. Punch, S., & McIntosh, I. (2003). The dynamics of giving and exchange between siblings. Paper presented at Connections that Count: An International Conference on Young people, Social Capital and Empowerment, 11–14 September 2003. Glasgow: University of Strathclyde. Raffaelli, M. (1992). Sibling conflict in early adolescence. Journal of Marriage and the Family, 54, 652–663. Roffey, S., Tarrant, T., & Majors, K. (1994). Young friends: Schools and friendship. London: Cassell. Scott, J. (2001). Power. Cambridge: Polity. Smart, B. (1985). Michel Foucault. London: Tavistock. Thomas, N. (2000). Children, family and the state. Basingstoke: Macmillan. Valentine, G. (1999). Oh please, mum. Oh please, dad”: Negotiating children’s spatial boundaries. In: L. McKie, S. Bowlby & S. Gregory (Eds), Gender, Power and the Household (pp. 137–154). Basingstoke: Macmillan. Weber, M. (1968). Economy and society (Vol. 3). New York: Bedminster Press. Westwood, S. (2002). Power and the social. London: Routledge.

A TALE OF TWO CITIES: HEALTH-COMPROMISING BEHAVIORS BETWEEN HUNGARIAN AND AMERICAN YOUTH Kevin M. Fitzpatrick, Bettina F. Piko and Darlene R. Wright INTRODUCTION Adolescents, because of the unique developmental stage they occupy, are particularly sensitive to their socio-cultural environment. Adolescents often define behaviors in light of prevailing attitudes, values, and norms (i.e. culture) established across primary social domains. Specifically, overarching social structures (e.g. economic, political, religious, etc.), working through the local landscape (e.g. neighborhood, school, peer networks, and family), play a vital role in shaping adolescent development and influencing psychological, behavioral, and social outcomes (Arnett & Arnett-Jensen, 1994; Greenberger et al., 2000; Grob et al., 1996; McArdle et al., 2000). For youth, definitions of normative behavior vary, yet socio-cultural context continues to be important in defining who they are and what they do. Culture defines accepted standards of behavior, lifestyles, and life chances. As such, socio-cultural influences have been particularly strong predictors of health-compromising behaviors for this population subgroup (Fitzpatrick, 1997; Fitzpatrick & LaGory, 2000; Gibbons et al., 1995; Graham et al., 1991). Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 189–212 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10010-X

189

190

KEVIN M. FITZPATRICK ET AL.

During adolescence, changes in social networks, a push for independence, increased substance abuse and sexual activity, and heightened aggression have common markers of developmental progression towards adulthood (Baer & Bray, 1999; Piko, 1998, 2001). Recent research suggests that because healthcompromising behavior may be influenced by specific societal circumstances, cross-cultural comparisons of youth may allow a better understanding of the complex relationship between context and behavior. In addition, such comparisons allow us to consider which has the greater impact on youth’s health-compromising behavior: the broader socio-cultural environment or the specific developmental concept of risk, which is more or less independent of culture (Arnett & ArnettJensen, 1994; Grob et al., 1996; Schneider, 1998). In this way, culture, or learned beliefs and practices, manifests itself via social mechanisms that either reinforce or sanction behaviors. Peers, parents, and teachers all influence how these behaviors are defined and redefined by an adolescent. Differences in parenting styles, family structure, parental attitudes, and peer relations have been noted in previous studies as important determinants of healthcompromising behavior among youth (Arnett & Arnett-Jensen, 1994; Fitzpatrick & LaGory, 2000; Grob et al., 1996; McCardle et al., 2000). Whether American or not, adolescents engage in such behaviors, in part, as a function of both individual attributes and a varied set of socio-cultural factors such as those discussed earlier. Cultural differences (e.g. beliefs about smoking and drinking; risk perception; values, norms, and social acceptance of certain behaviors) clearly influence health-compromising behaviors (Gibbons et al., 1995; Johnson & Johnson, 1999; McArdle et al., 2000). Understanding the role that risk factors play in predicting such behavior, as well as identifying which protective factors help mediate risk, has been a crucial aspect of worldwide health promotion for adolescents in the 20th century. Not specific to any culture but rather to youth in general, risk and protective factors have become a salient combination for examining cross-cultural differences in health-compromising behavior.

RISK AND PROTECTIVE FACTORS: CORRELATES OF HEALTH-COMPROMISING BEHAVIOR Risk and protective factors were originally introduced to help understand healthcompromising behaviors among youth by combining social developmental theory (Hawkins et al., 1992) with social control and social learning theories (Jessor, 1993, 1998). Rather than conceptualizing behavior as a function of socially determined or individual choices, the focus of this approach has been to identify factors that promote or discourage health-compromising behavior. Examining multiple

A Tale of Two Cities

191

domains of risk and protection (individual, peer, family, school, and community), these factors provide a means for disentangling the effects of certain contexts on behavioral outcomes. In many cases, research on health-compromising behavior has found considerable foundational support from theories in both sociology (Gottfredson & Hirschi, 1990; Sampson & Laub, 1993), and psychology (Jessor, 1998; Jessor & Jessor, 1977) that argue certain behaviors tend to cluster themselves within certain socio-demographic groups (i.e. age, race, gender, class, etc.) (Brener & Collins, 1998; DuRant et al., 1999; Warrren et al., 1997). The idea that behaviors replicate themselves and are reinforced by certain environmental circumstances (neighborhood structure, cultural values, peer attitudes, etc.) is a general assumption that underlies our interest in examining health-compromising behaviors across cultures. Whether a function of social ecology, developmental patterns, or health behavior clustering, a body of empirical literature points to the generality of certain types of negative behaviors and the common markers that often are used to predict these behaviors (e.g. Dryfoos, 1990; Jessor, 1993, 1998). We know that risk is not simply the absence or opposite of protection; rather risk is seen as a high level of increased susceptibility to negative behavioral outcomes. Protection has been identified as the “positive counterpart of risk” (Masten, 1994). Yet clearly, more than just the opposite of risk, it is associated with a reduced incidence of negative behavioral outcomes. Protective factors are used to manage, or adapt to, risk vulnerability – to shield youth from negative circumstances that hinder successful development while fostering resiliency and the likelihood of more positive outcomes (Hawkins et al., 1992). Again, risk and protection are not simply at opposite ends of a single dimension but, in fact, are conceptually distinct in the way they influence outcomes independently of one another. While their interaction may be important, the current study focuses only on their independent effects in predicting negative behavioral outcomes among a sample of AfricanAmerican and Hungarian adolescents. There are a number of socio-demographic factors that have been suggested as important correlates of health-compromising behavior among youth. For example, research shows that boys tend to engage in more substance abuse and aggressive behavior compared to girls (Fitzpatrick, 1997; Johnson & Hoffmann, 2000; Loeber et al., 1998). Moreover, older adolescents engage in more health-compromising behavior than younger adolescents (e.g. Fitzpatrick, 1997; Kelley et al., 1997). Beyond these standard socio-demographics, several individual risk factors are predictive of negative behavioral outcomes including psychological symptomatology (i.e. depression), as well as youth’s favorable attitudes towards taking risks, particularly drug and alcohol use (e.g. Flisher et al., 2000; Reinherz

192

KEVIN M. FITZPATRICK ET AL.

et al., 2000) Individual protective factors include a positive self-image (Dumont & Provost, 1999; Hawkins et al., 1992; Jessor, 1998) and high academic achievement reflecting higher intellectual capacity and lower probability of healthcompromising behaviors (Dekovic, 1999; Masten, 1994). Grades while embedded within the school environment are seen as a product of intellectual capacity as much if not more than the learning environment itself. Another important set of risk and protective factors are found in the family domain. Risk factors include child mistreatment (Hawkins et al., 2000; Loeber et al., 1998; Smith & Thornberry, 1995) and disruption or change in the family routine or structure due to separation, divorce, or death of a parent (Hawkins et al., 2000; Henry et al., 1996; McCord & Ensminger, 1995). Protective family factors include parental monitoring, in the form of setting curfews, knowing where children go with friends, and eating evening meals together. Regardless of cultural differences, research findings suggest that positive parental interaction protects youth against risk and enhances resiliency (Arnett & Arnett-Jensen, 1994; Dekovic, 1999; Hawkins et al., 1992). Talking with parents about problems is one form of bonding related to the quality of the attachment between parent and child that also protects against negative behavior (Dekovic, 1999; Fitzpatrick, 1997). Peer-related risk and protective factors are also important predictors of healthcompromising behavior among youth. Socializing with peers who engage in delinquent behavior or who have more positive attitudes towards delinquency greatly increase one’s chances of similar risk behaviors (Elliot, 1994; Piko, 2000). Likewise, the nature and quality of peer group relations can serve as important protective mechanisms for youth. Peer influences, while often viewed as negative, do provide support and resistance from engaging in negative behaviors. While controversial, some research has found that membership in school sports teams or school clubs offer ways to channel energies in socially appropriate ways with peers who are more likely to engage in socially acceptable forms of behavior (Dumont & Provost, 1999; Holland & Andre, 1987). A final set of risk and protective factors are in the school domain. Lack of bonding with the school and a feeling that no one cares or provides individual attention have been noted in earlier studies as important risk correlates of healthcompromising behavior (Fitzpatrick, 1997; Hawkins, 1995; Maguin et al., 1995). As suggested earlier, a student’s negative perception of how much people cared or paid attention to them at school was a potentially important risk factor contributing to a lack of bonding at the school level. On the other hand, students who talk with their teachers about problems may actually be using them as a form of support and, while the better students are more likely to seek these teachers out, that relationship nevertheless may be important to insulate students from negative school circumstances that might encourage health-compromising behavior.

A Tale of Two Cities

193

This paper seeks to describe two seemingly different populations of youth from American and Hungarian samples, with emphasis on several questions: (1) What are the general socio-cultural characteristics of Hungary and the United States and what role might they play in understanding health-compromising behavior among youth? (2) What differences or similarities in risk exist between youth in the two samples? (3) Are there any specific protective factors that are important to predicting health-compromising behavior in the two samples; in short, is protection universal or is it culturally specific?

CULTURE OF RISK? We know at the outset there are considerable differences between the two countries of interest, i.e. American and Eastern European (Hungarian). Disparities exist between the United States and many European countries in terms of the use of transportation, mobility, legal obligations, financial responsibilities, and risk behavior (Arnett & Arnett-Jensen, 1994). European youth generally are more integrated into community life and have regular contact with extended family members (e.g. Arnett & Arnett-Jensen, 1994; Gibbons et al., 1995). In addition, different parenting styles and attitudes, as well as attitudes toward peer relations prevail across these two samples. These differences seem to suggest that any comparison between the United States and Hungary would be striking. Knowing that differences exist, the question remains: is health-compromising behavior such among adolescents that it supersedes specific cultural context? Trying to describe differences between these sampled adolescents and speculating why they exist is the underlying purpose of this present study. Why has Hungary been selected for this cross-cultural comparative study? In Eastern Europe, Hungary is unique in several ways, particularly with regard to the health condition of its general population (Cockerham, 1999). Since the 1960’s, socialism was more pragmatic than in many other socialist countries and Hungary enjoyed a more successful economic development than the majority of its neighbors. Yet, despite the relative wealth of the country, the health conditions of the population did not mirror this economic upswing. The mortality rates were among the worst in socialist countries and the occurrence of smoking and drinking were as high as any rates in all of Eastern Europe. As a result, Hungary experienced changes that introduced a shared value system with, at best, suspect results (Piko & Fitzpatrick, 2001). While several recent studies have examined cross-cultural differences in behavioral outcomes among adolescents, none to our knowledge has compared specific health-compromising behaviors between samples of American and Hungarian adolescents. In addition to the absence of such data, we argue that this

194

KEVIN M. FITZPATRICK ET AL.

type of examination is vital to understanding the role of context in determining behavioral outcomes for adolescents in two very different countries. Hungary has been undergoing significant political and economic post-socialism transformation for more than a decade. With its current socio-cultural orientation and political expectations, this examination appears to be a timely one. An examination of crime data between the two countries reinforces the notion that these two countries are different. Whether using serious crimes (U.S. in 2000 had nearly 13,000 murders compared to Hungary with 200) or drug offenses as indicators (U.S. in 2000 there were 560 per 100,000 persons compared to Hungary where there were 34 per 100,000 persons), the criminal structure between these two social systems is drastically different. This is also true when examining crime and victimization rates among youth where in the United States, victimization is five to six times higher than most European countries (Rapid Intelligence, 2004). Additionally, we find obvious racial differences between the two samples. The American sample is comprised of African-American youth living in the Southeastern United States. Is it generalizable to youth outside of the South, to youth other than African-Americans? A body of research supports the general assumption that race is an important social structural characteristic important in determining health-related outcomes (Andrulis, 1987; Fitzpatrick & LaGory, 2000; LeClere et al., 1997; Williams, 1996). This research also points out the importance of socioeconomic distinctions and their role in determining health and risk outcomes, particularly among youth (e.g. Elliot, Hagan & McCord, 1998; Fitzpatrick & LaGory, 2000; Kessler & Neighbors, 1986; Levine & Rosich, 1996). Thus, while the desirable comparison would be to have similar racial compositions in both samples, the fact that the communities from which both sets of youth are drawn, have nearly one-third living at or below an established level of poverty may be an important commonness to consider as we look at the differences and similarities regarding what factors predict health-compromising behavior among these two groups. Finally, while the two samples are divergent, the somewhat underdeveloped nature of the two communities should provide for a unique comparison between youth from “East and West” and the role that context plays in understanding health-compromising behavior among youth.

DATA AND METHODS American sample. The sample consists of 1,538 African-American middle and high school students from eight schools in Bessemer, Alabama. Bessemer city is located on the outskirts of Birmingham with a population of nearly 50,000 people. The majority of residents living in the city are African-American (60%),

A Tale of Two Cities

195

over one-third (35%) are from female-headed households with children under the age of 18, and over 30% of the residents are living below the poverty level (U.S. Bureau of Census, 1993). Of the sampled students, approximately 49% were males; the median age of the sample was 14 years of age (9th grade). The 2001–2002 average daily attendance (ADA) for the middle and high schools in the system was 2,028 students. Student participation in the study was voluntary, leaving us with a response rate of approximately 76%. A letter of consent was sent home by the school district explaining the purpose of the health assessment. If students did not want to participate, they were to return the letter signed by a parent/guardian. The 24% who did not participate consisted primarily of youth whose parents refused to allow them to participate (15%), and those students who were absent, suspended, or no longer attending school in the system. If looking at just refusals, the response rate is similar (91%) to that reported in the Hungarian sample. It should also be noted that because some proportion of those students who were not included in the sample were unavailable due to behavioral reasons, the data reported here may actually slightly under-represent the true picture of what is taking place in this particular school district. Hungarian sample. This sample consists of 1,242 middle and high school students from seven schools in Szeged, southern Hungary where data collection was based on random selection of classes. Szeged is the third largest city in the country and a major urban center in the southeast of Hungary with a population of approximately 180,000 people. The majority of residents living in Szeged is Caucasian (90%) and from two-parent families (75%). Approximately 30% live at or below an established standard of poverty, based on Hungarian income levels. Of the sampled students, 53% were males; the median age of the sample was 16 years of age (10th grade). Of the 1,500 questionnaires given to students, 1,242 were returned, giving us a response rate of about 84%.

Procedure The data for the Hungarian sample were collected in the spring of 2000, with the American sample collected in spring 2001. The instruments contained questions on socio-demographics, family structure, risk and protective factors, mental health status, student’s attitudes towards school, and student’s assessments of their relationships with teachers and parents. In both cases, the questionnaires were selfadministered under close supervision by classroom teachers or student teachers. In both samples, school system IRB regulated consent forms were required to be signed by parents for those students under the age of 14. Teachers did not have any primary role in the data collection; they were only present to help monitor

196

KEVIN M. FITZPATRICK ET AL.

the classroom and answer any questions if necessary. Students were reminded that their responses would remain confidential.

Measurement Health-compromising behavior. The measure used to assess health-compromising behavior among adolescents from both samples was a composite measure based on youth’s response to questions about cigarette smoking, drug and alcohol consumption, carrying weapons in school, and fighting behavior. The questions were asked in identical format, albeit specific for each risk; the items are comparable and allowed construction of this type of composite measure (Group for the Advancement of Psychiatry, 1999). We asked, “In the last 30 days, how much have you . . .” smoked cigarettes, drank alcohol, smoked marijuana, took illicit drugs (i.e. amphetamines, barbituates, etc.), and sniffed glue or other aerosols. In addition, we asked: “In the last 30 days how often did you carry a weapon(s) to school?” and “In the last 30 days how often were you in a physical fight with another person?” The scores were summed for each of the indices and, in the American sample, the composite health-compromising behavior measure had a range of 0–22 and was reliable with a Cronbach’s alpha of 0.71. For the Hungarian sample, the composite health-compromising behavior measure also had a range of 0–22 and was reliable with a Cronbach’s alpha of 0.69. Socio-demographics. There are two variables that we control for in the analysis: gender (1 = female) and age (years) are included in this present study. Risk factors. Depressive symptomatology is an important individual risk for health compromising behaviors (Green & Pope, 2000). In the American sample, depression is measured using a shortened version of the original 20-item Center for Epidemiological Studies for Depression (CES-D) scale designed to index affective depressive symptoms (Radloff, 1977). The present study uses a brief version of the original CES-D; eight items with scores ranging from 0 to 24. The eight questions focus mostly on the depressive affect domain of the CES-D. In addition to asking questions about whether students felt sad, lonely, or fearful, other questions asked students if they had difficulty sleeping, if their life felt like a failure, if they had trouble getting along with others, etc. The formal properties of the original CES-D have been demonstrated elsewhere with very high coefficients of internal consistency when administered to adolescents (e.g. Garrison et al., 1991; Prescott et al., 1998). Among the American sample of students, the brief CES-D was reliable with a Cronbach’s alpha of 0.85. In the case of the Hungarian sample, depression was measured using a shortened version of the original 27-item Children’s Depression Inventory (CDI). The CDI

A Tale of Two Cities

197

is a self-rated depressive symptom scale for children and adolescents adapted from the Beck Depression Inventory for adults (Kovacs, 1983, 1992). Previous data demonstrate acceptable psychometric properties for the original scale and its shortened version (Smucker et al., 1986). Each item of the original and shortened version assesses a single symptom, such as sadness, fear, etc. and is coded 0 to 2. In the present study, no cutoff scores were used to develop specific diagnostic cutoff criteria. Again, we were interested only in the summated symptomatology and its role in predicting health-compromising behavior. The CDI, both the original and the shortened (eight items based on the factor structure of the original scale) versions, are widely used in psychiatric care and epidemiological research in Hungary and the United States (Masi et al., 2001; R´ozsa et al., 1999; Way et al., 1994). The shortened version of the CDI, based on the present data, was reliable with a Cronbach’s alpha of 0.75. It is important to note that several studies document high intercorrelations between these two measures of symptomatology (CES-D and CDI) that are often used interchangeably (Charman & Pervova, 1996; Chartier & Lassen, 1994; Garrison et al., 1991; Prescott et al., 1998). As such, we anticipate no problems with using these two different measures to assess the presence of depressive symptoms among American and Hungarian youth. A second individual risk factor was the student’s attitudes towards drug and alcohol use. The variable asked students how they personally felt about using drugs and alcohol with the categories for this ordinal variable ranging from 1 = strongly disagree to 5 = strongly agree. Two family risk factors were included in the analysis. A family violence variable that asked students if they were ever slapped, punched, kicked, or beaten by their parent(s) or guardian. The ordinal variable had response categories where 0 = never, 1 = once or twice, 2 = several times, 3 = most of the time, and 4 = all of the time. Another family risk factor was family structure, specifically whether or not youth currently lived in a non-intact family where at least one or both parents were missing (Fitzpatrick, 1993). The family structure variable was coded 0 = both parents present in the home, 1 = mother or father in the home where one is a stepparent, 2 = only mother or father in the home, and 3 = neither biological parents in the home. Two variables were used to assess peer risk factors. Students were asked, “Are you a member of a gang?” While the exact wording of this question was different between the two samples, similar characterizations were used for responses relevant to each of the cultures. This dichotomized variable (1 = yes) was designed to assess deviant peer group membership and the potential pressure for youth to engage in more extreme forms of health-compromising behaviors (i.e. fighting, illicit drug use, weapon possession, etc.). A second peer risk factor assessed friend’s attitudes towards drug and alcohol use which again was an ordinal variable with response categories ranging from 1 = strongly disagree to 5 = strongly agree.

198

KEVIN M. FITZPATRICK ET AL.

A final risk factor assessed student’s perception of how little attention they received from their teachers. The ordinal variable asked, “How often do you feel that you get personal attention from your teachers?” was coded 1 = teachers pay attention to me all the time, 2 = most of the time, 3 = some of the time, 4 = hardly ever, and 5 = teachers never pay attention to me. Protective factors. Protective factors were viewed as individual, family, peer, and school-related assets that foster competency, promote successful development, and build resiliency in youth. One important individual protective factor is self-esteem. In both samples, we use Rosenberg’s 10-item self-esteem scale (Rosenberg, 1965, 1979). These 10 items asked students questions pertaining to how they perceive themselves. Answers range from 1 = strongly agree to 4 = strongly disagree. The self-esteem measure for the American sample was moderately reliable with a Cronbach’s alpha of 0.67. Likewise, for the Hungarian sample, self-esteem was moderately reliable with a Cronbach’s alpha of 0.66. Another individual protective factor, academic achievement, is a subjective assessment by students of the “grades they mostly receive in school” ranging from 1 = mostly D’s and F’s to 7 = mostly A’s. Two family domain protective factors were used in the analysis. One was how often youth talk to their parents or guardian(s) about their personal problems. Family talk was an ordinal variable where 0 = never talk with my parents, 1 = hardly ever talk to my parents, 2 = sometimes talk to my parents, 3 = most of the time talk to my parents, and 4 = talk to my parents about my personal problems all of the time. We assessed parental monitoring using an ordinal variable that asked students how often parents imposed a curfew ranging from 0 = never, 1 = hardly ever, 2 = sometimes, 3 = most of the time, and 4 = all of the time. One peer related protective factor was the number of clubs/organizations a student belonged to in school, it was included as an interval-level variable by simply asking students how many school clubs or organizations were they a member of last year. Finally, we know that because youth spend a great deal of time in their schools, school personnel are often considered a vital link in successful adolescent development. We asked students a similar question about their teachers as we did their parents: “How often do you talk with your teachers about your personal problems?” Again this ordinal variable ranged from 0 = never talk to my teacher about personal problems to 4 = talk to my teacher all the time about my personal problems.

Analysis Data analysis used SPSS for Windows 11.0 with minimum significance levels for all statistical tests of p < 0.05. Beyond the simple descriptive statistics found

A Tale of Two Cities

199

Table 1. Descriptive Statistics for Samples of American (n = 1,538) and Hungarian (n = 1,240) Youth. Variablesa

American %

Mean

Health-compromising index Sociodemographics Gender (female) Age (years)

Hungarian S. D.

%

Cramer’s V

Mean

S. D.

2.41b

2.78

3.30

3.99

13.48

2.19

15.65b

2.11

7.77

6.92

2.93b

2.42

49.7

46.9

Risk factors CES-D Scale/CDI Own attitudes towards drugs/alcohol Strongly disagree 59.2 Disagree 14.0 Don’t care 16.7 Agree 3.7 Strongly agree 3.5

72.8b 11.0 10.4 2.6 1.5

Hit by parent(s) Never One or two times Several times Most of the time All of the time

71.7 14.8 6.3 2.0 1.5

69.0b 19.3 8.6 1.5 1.0

0.08

Family structure Both parents One parent and stepparent Mother or father No parents

28.2 12.5 45.7 13.8

66.2b 10.8 20.2 2.5

0.41

52.4b 18.4 17.1 6.3 3.5

0.14

14.3

22.9b

0.16

11.4 14.4 33.0 12.5 27.8

3.0b 8.9 32.7 30.8 23.5

0.22

Friends attitudes towards drugs/alcohol Strongly disagree 50.4 Disagree 11.0 Don’t care 25.5 Agree 5.7 Strongly agree 5.5 Gang membership (Yes) Teacher inattention All the time Most of the time Sometimes Hardly ever Never

0.35

200

KEVIN M. FITZPATRICK ET AL.

Table 1. (Continued ) Variablesa

American %

Protective factors Self-esteem

Hungarian

Mean

S. D.

17.90

5.11

%

Cramer’s V

Mean

S. D.

27.08b

4.72

Grades Mostly D’s and F’s Mostly C’s and D’s Mostly C’s Mostly B’s and C’s Mostly B’s Mostly B’s and A’s Mostly A’s

5.9 18.1 6.8 32.1 4.4 26.7 5.3

2.6b 24.3 15.6 30.5 9.3 15.9 1.2

0.20

Curfew set Never Hardly ever Sometimes Most of the time All of the time

16.0 10.0 24.1 14.8 35.1

22.1b 19.0 12.6 18.9 26.3

0.21

Family talk Never talk Hardly ever talk Sometimes talk Most of the time talk All of the time talk

13.6 13.7 37.3 18.7 16.9

8.5b 21.3 29.1 27.9 12.8

0.24

School club membership Teacher talk Never talk Hardly ever talk Sometimes talk Most of the time talk All of the time talk a Due

1.00 44.5 20.9 24.9 6.1 3.2

1.53

1.00 43.6b 35.8 14.4 4.2 1.5

1.44 0.19

to some missing cases and rounding, percentages do not add up to 100%. ␹2 difference in proportions/difference in means test.

b p < 0.001;

in Table 1, we examine the differences in the frequency of health-compromising behaviors between the American and Hungarian sample. In both Tables 1 and 2, we used a test of proportional differences using ␹2 . In addition to testing for significance, Cramer’s V was added as a measure of the strength of association, i.e. how strong is the difference in risk and protective factors and health-compromising behavior between the two samples. Cramer’s V ranges from 0 to 1 and is

A Tale of Two Cities

201

Table 2. Frequency of Health-Compromising Behaviors Among American and Hungarian Youth in the Last 30 days. Health-Compromising Behaviorsa

American (%)

Hungarian (%)

Cramer’s V

cigarettesb

Smoke Never smoke cigarettes <1 per day 1–5 per day 6–10 per day 11–20 per day

80.1 16.4 1.3 1.0 1.2

49.0d 18.2 11.4 12.4 9.1

0.42

Drink alcohol Never drink alcohol 1–2 drinks 3–9 drinks 10–19 drinks 20–39 drinks 40+ drinks

73.7 12.9 8.3 2.6 1.0 1.2

46.8d 34.1 15.1 2.3 1.4 1.0

0.26

Smoke marijuana Never smoke marijuana 1–2 times 3–9 times 10–19 times 20–39 times 40+ times

84.8 5.2 3.8 1.8 1.3 1.8

93.5d 3.4 1.7 0.6 0.5 0.4

0.16

Take illicit drugs (Coke, Speed, LSD) Never take illicit drugs 1–2 times 3–9 times 10–19 times 20–39 times 40+ times

96.2 1.1 0.8 0.7 0.3 0.5

98.2c 1.2 0.1 0.2 0.1 0.2

0.07

Sniff glue or aerosols Never sniff glue or aerosols 1–2 times 3–9 times

93.1 3.0 3.9

99.0d 0.7 0.3

0.11

Carry weapons Never carry weapons 1 time 2–3 times 4–5 times 6+ times

70.2 13.0 10.1 3.6 2.6

75.2d 7.8 5.5 2.3 9.1

0.13

Get in physical fights Never fight 1 time

33.3 19.3

57.2d 19.5

0.27

202

KEVIN M. FITZPATRICK ET AL.

Table 2. (Continued ) Health-Compromising Behaviorsa 2–3 times 4–5 times 6+ times

American (%) 21.5 8.2 16.6

Hungarian (%)

Cramer’s V

12.4 3.0 7.2

a Due

to some missing cases and rounding, percentages do not add to 100%. difference in proportions test. c p < 0.05. d p < 0.01. b ␹2

interpreted similarly to that of a bivariate correlation. Finally, we examine the multivariate relationships among risk and protective factors using OLS regression. The regression results are presented in three blocks: socio-demographic, risk factors, and protective factors. In addition to being interested in the contribution individual variables make to the overall explained variance, we were also interested in these three blocks and their contribution to the overall R2 in the models.

RESULTS Table 1 presents descriptive statistics for the variables used in the present analysis. While the results show that in both samples youth engage in health compromising behavior, youth in the American sample, on average, do so more than youth in the Hungarian sample. The samples are similar in the percentage of males and females, with the Hungarian sample slightly older than the American one. Both sets of youth report moderate levels of both self-esteem and depressive symptomatology, yet these along with all of the other variables except sports team membership are significantly different. Interestingly, the Hungarian sample had nearly 10% more students reporting gang membership compared to the African-American sample. Youth gang membership appears to mean something different in this culture than what may typically be thought of in American culture; however, Hungarian youth do report that while these gangs may not be as inherently violent, they still increase risk-taking and membership is often selective, requiring some initiation through other forms of health-compromising behaviors. In addition to the differences in peer groups, Hungarian family structure for this set of samples is different, with nearly three times the number of Hungarian youth reporting living with both parents. These same youth tend to talk more with parents about their problems than their American counterparts.

A Tale of Two Cities

203

Table 2 presents percentage differences in health-compromising behavior between adolescents in the American and Hungarian samples. The statistical significance of those differences is calculated using Chi-square at both 0.05 and 0.01 levels of significance. In addition, Cramer’s V, a measure of strength of the association, is also presented. Upon close inspection of this table, several dramatic differences are apparent. First, as expected given the cultural environment in Europe generally and Eastern Europe specifically, Hungarian youth smoke cigarettes significantly (p < 0.01) more than American youth. While this is again part of an accepted European standard among youth, it is highly problematic with major implications for adult health-compromising behaviors and lifestyles. Also, twice as many Hungarian youth in the sample report drinking at moderate levels compared to American youth; the percentages reporting drinking at higher levels are not as large, nevertheless significant (p < 0.01). The differences between the two samples for other drug risk behaviors such as smoking marijuana, taking other illicit drugs (amphetamines, LSD, cocaine, etc.), and sniffing aerosols are smaller but still statistically significant. Americans engage in these behaviors more often than Hungarians. Another striking difference between the samples of American and Hungarian youth is the percentage differences in the reporting of fighting and weapon carrying behavior (p < 0.01). In some cases, American youth fight and carry weapons two to three times more than Hungarian youth. Considering gang membership differences found in Table 1, there is much less violent behavior taking place among these groups in Europe. Conversely, American youth gangs typically engage in more violent and illegal substance abusing behaviors. Table 3 presents the results of the composite health-compromising behavior measure regressed on a set of socio-demographic, risk, and protection variables for African-American and Hungarian youth. In the first three columns of Table 3 there are unstandardized and standardized regression coefficients for a set of hierarchical models that examine these relationships among American youth. In columns 4–6 there are a similar set of regression results for Hungarian youth. The first model examines the socio-demographic/risk variables. The model R2 was significant and, in both samples, older youth and males engaged in more healthcompromising behavior. The second model introduces individual, family, peer and school risk factors. For the American and Hungarian models (2) we find strikingly similar results; i.e. all of the risk factors are in the expected direction (positive) and significant in predicting health-compromising behavior with the exception of family structure in the American sample and friend’s attitudes towards drugs and alcohol in the Hungarian sample. Generally, we find that youth reporting more depressive symptomatology; greater frequency of being hit, punched, slapped or kicked by a parent or guardian; gang membership; more favorable attitudes toward

204

KEVIN M. FITZPATRICK ET AL.

Table 3. Health-Compromising Regressions for Samples of American (n = 1,538) and Hungarian (n = 1,240) Youth. Variables

American

Sociodemographics Gender (Female = 1) Age

(2)

(3)

(1)

(2)

(3)

␤ (b)

␤ (b)

␤ (b)

␤ (b)

␤ (b)

␤ (b)

−0.21 −1.62b 0.20 0.37b

Risk factors CESD/CDI Own drug/alcohol attitudes Hit by parent/guardian Family structure Friends drug/alcohol attitudes Gang membership Teacher inattention

−0.14 −1.08b 0.06 0.09b

−0.14 −1.00b 0.01 0.01

0.06 0.01b 0.31 1.17b 0.15 0.86b −0.01 0.004 0.07 0.23b 0.31 3.42b 0.05 0.007

0.05 0.01b 0.31 1.14b 0.15 0.85b −0.01 0.005 0.06 0.19b 0.28 3.17b 0.01 0.003

Protective factors Self-esteem

Curfew set by parents Family talk School club membership Teacher talk 0.93 0.09

a p < 0.05. b p < 0.01 c p < 0.01

−0.32 −1.77b 0.15 0.20b

−0.27 −1.52b 0.07 0.09b

−0.23 1.29b 0.01 0.07

0.07 0.09b 0.29 0.93b 0.05 0.15a 0.04 0.13a 0.04 0.09 0.23 1.45b 0.07 0.12b

0.09 0.09b 0.29 0.91b 0.04 0.15a 0.04 0.13a 0.04 0.09 0.23 1.50b 0.06 0.10b

0.18 0.04 −0.08 −0.19b −0.10 −0.27b −0.02 −0.004 −0.02 −0.06 −0.02 −0.05

Grades in school

Constant R2

Hungarian

(1)

one-tailed t-test. unadjusted R2 hierarchical F-test.

1.97 0.41c

2.79 0.44c

0.06 0.06a −0.11 −0.11a −0.11 −0.19b −0.03 −0.08 0.02 0.01 −0.02 −0.06 0.15 0.12

0.93 0.34c

0.87 0.36c

A Tale of Two Cities

205

drug and alcohol use, and a lack of attention from their teachers are more likely to be frequent risk takers. The third model, controlling for these sociodemographics and risk factors, adds a set of protective factors. In both models, the R2 significantly increased with the addition of these factors; several protective factors were consistent predictors for health-compromising behavior. Sampled youth with both higher grades and parents who set curfews report fewer health-compromising behaviors. Club membership, talking with a teacher/parent were protective factors in the right direction but did not achieve statistical significance. One interesting finding to note in this group of protective variables was the effect and significance of the selfesteem variable. It would be reasonable to expect self-esteem to have a negative effect on health-compromising behavior (i.e. youth with a better, more secure sense of themselves will engage in fewer health-compromising behaviors). This was not the case in the current study; in fact, self-esteem was positive and, in the Hungarian sample, significant. Of course, given the cross-sectional nature of our data, we are unable to draw any direct causal conclusions from this examination, but this relationship appears to be an important one in need of further investigation. The relative consistency between the second and third regression model suggests that protective factors appear to have a small additive effect but are not playing a mediating role by significantly reducing or changing the negative effects of risks on health-compromising behavior. One final note about the models is their R2 ’s. In both samples the final R2 was unusually high; over 40% of health-compromising behavior explained by a set of risk and protective factors. This once again points to the similarity in our findings between samples and the strength of our earlier arguments regarding the generality that may emerge beyond any specific findings.

DISCUSSION AND CONCLUSIONS The primary focus of this paper has been to examine the applicability of risk and protective factors for understanding the similarities and differences in more extreme forms of health-compromising behavior among samples of American and Hungarian youth. Using two samples of primarily middle- and high-school aged youth attending schools in moderately high-risk settings, the models do a good job in predicting health-compromising behavior. Again, results underscore not only the importance of using this type of framework for understanding adolescent behavior, but also how similar the results are when considering the role of risk and protection in predicting extreme health-compromising behavior among youth in two very different countries.

206

KEVIN M. FITZPATRICK ET AL.

Both American males and Hungarian males consistently engage in more healthcompromising behavior than do their female counterparts. Earlier studies have shown there were no differences in smoking and drinking among Hungarian boys and girls (Piko, 2000); however, we find that in the case of the composite risk index, there were gender differences in both samples. Additionally, we find older youth, those suffering from psychopathological symptoms, exposed to violence in the home; with more positive attitudes towards drug and alcohol use engaging in more health-compromising behavior. As suggested earlier, risky health behaviors take place in multiple domains and the compounding of exposure to these risks additively has considerable impact on both samples of youth. Keeping youth insulated from the negative effects of these risks is a major worldwide public health challenge in the 21st century. While there are certain individual protective factors that we know help mediate exposure to negative circumstances, the role of parents, teachers, and community leaders has been noted elsewhere to be just as important (Cairns & Cairns, 1994; Dryfoos, 1990; Jessor, 1998). While this is often a futile effort in some settings, our results indicate that family may play some role in protecting youth from engaging in more healthcompromising behavior, independent of the risks to which they are exposed. Being a resource for adolescents either through monitoring or listening is an important role for parents to be playing in any country. Clearly, our results suggest that parenting styles, while unique across both samples, have some universal implications. Finally, self-esteem did not operate as a typical protective factor; specifically, greater self-esteem creates a sense of self-worth that helps youth make more informed decisions about their health and behavior and typically engage in fewer health-compromising behaviors (Hawkins, Catalano & Miller, 1992). However, while there appears to be some controversy surrounding this issue (e.g. Petraitis, Flay, Miller, Torpy & Greiner, 1998), we found that, in both samples, higher self-esteem scores were related, although not always significantly, to more healthcompromising behavior. It could be that those youth with a greater sense of selfworth and control feel like they can make these decisions without any serious negative implications. As we proposed earlier, this relationship needs further investigation. Finally, as noted earlier, the racial composition of the American sample may be of some importance when considering the overall generalizability of our findings. Certainly, a different American sample of youth could yield different results. Nevertheless, we believe that these two samples, while not racially matched, are comparable in other ways that make this cross-cultural analysis a valid one. While we believe this current research to be extremely important to developing more systematic approaches to comparative analyses of adolescent

A Tale of Two Cities

207

health-compromising behavior, there are some important limitations that should be noted. The data used here to examine comparative health-compromising behaviors are cross-sectional; in order to establish causality, longitudinal data is necessary. As we have seen, some of these relationships may require refinement with regard to respecification (changing the causal sequencing) of relationships. Nevertheless, exploratory analyses have found some interesting differences and similarities between the two samples. We should also note that these samples were drawn independent of one another and while the instruments were matched with a good deal of precision, the sampling framework used to develop the samples was different. In addition, we examined only one culture unique from that of the United States, thus making any conclusion that culture does not matter in determining health-compromising behavior is a tenuous one. We also know that there is variation in response styles and that those differences may often create artificial group differences (Gibbons, Larsen & Garrard, 1995; Hui & Triandis, 1989). Thus, these samples of Hungarian and American youth may define things like gang membership, sadness, and weapons differently, yet the results are surprisingly consistent across groups. As we mentioned earlier, these two samples are not necessarily representative of the entire Hungarian and American youth populations. The American youth are black, lower class and the Hungarian youth are white and lower-middle class. Nevertheless, the comparison makes an important point regarding understanding differences across cultures and the role of risk and protection in disentangling health-compromising behaviors. While we urge the reader to be cautious in drawing conclusions about the results, we believe that the findings seem to be suggesting an important leading conclusion: culture is important but the general behaviors of adolescents may, in some cases, supersede cultural distinctiveness. Of course there is no way to find a perfect match between Eastern European and American cultures, rather what we attempt to do is to locate two similar environmental contexts with similar economic conditions and examine the role of risk and protection among a similar age group in these two cultures. There is no perfect match and we make no claims that these two samples are representative of their countries – nevertheless this comparison is important in developing a more comprehensive analysis between these two cultures and their youth. These two disparate worlds of adolescents are sharply defined by their history, culture, current economic situation, and political underpinning. While we know that these structures have routinely dictated behaviors for young and old alike, the present study has found considerable similarity in what does and does not predict health-compromising behavior among African-American and Hungarian youth. We do not look at this study as the conclusive evidence that culture does not matter in determining health risk behavior among youth rather the provocative

208

KEVIN M. FITZPATRICK ET AL.

results provide important evidence to justify a more comprehensive analyses of cross-cultural health-compromising behavior. If anything, the results suggest an unusual amount of commonality regarding health-compromising behavior and its predictors. We know that the type of healthcompromising behavior between the two groups is different, yet the results indicate that the mechanisms impacting behavior in both a positive and negative way are strikingly similar. Does culture make a difference? Does place matter? With over 40% of the variation explained in health-compromising behavior among these samples of youth, the results of this preliminary analysis seem to suggest otherwise.

ACKNOWLEDGMENTS The authors would like to thank Chris Snead for her assistance with the Birmingham (AL) data collection phase of the project. In addition, we thank the Bessemer City School system and Mary Fitzpatrick, Safe and Drug-Free Schools coordinator, for making available to us their data used in this study. The authors also thank Mary Burg and Angela Dancso for their assistance in the data collection and processing phase of the Hungarian study. That part of the study was supported by the ETT T9 003/2000 grant of the Ministry of Health Care (Hungary) the Bolyai Research Fellowship provided to Dr. Piko through the Hungarian Academy of Sciences (Hungary), and a Fulbright Research Fellowship to Kevin Fitzpatrick.

REFERENCES Andrulis, D. P. (1987). The urban health penalty: New dimensions and directions in inner city health care. American College of Physicians Position Paper (no. 1). Washington, DC: National Public Health and Hospital Institute. Arnett, J. J., & Arnett-Jensen, L. (1994). Socialization and risk behavior in two countries: Denmark and the United States. Youth and Society, 26, 3–23. Baer, P. E., & Bray, J. H. (1999). Adolescent individuation and alcohol use. Journal of Studies on Alcohol, 13S, 52–62. Brener, N. D., & Collins, J. L. (1998). Co-occurrence of health-risk behaviors among adolescents in the United States. Journal of Adolescent Health, 22, 209–213. Cairns, R. B., & Cairns, B. D. (1994). Lifelines and risks: Pathways of youth in our time. Cambridge, UK: Cambridge University Press. Charman, T., & Pervova, I. (1996). Self-reported depressed mood in Russian and UK school children: a research note. Journal of Child Psychology and Psychiatry, 37, 879–883. Chartier, G. M., & Lassen, M. K. (1994). Adolescent depression: Children’s depression inventory norms, suicidal ideation, and (weak) gender effects. Adolescence, 29, 859–864.

A Tale of Two Cities

209

Cockerham, W. C. (1999). Health and social change in Russia and Eastern Europe. Routledge: New York. Dekovic, M. (1999). Risk and protective factors in the development of problem behavior during adolescence. Journal of Youth and Adolescence, 28, 667–685. Dryfoos, J. (1990). Adolescents at risk: Prevalence and prevention. New York: Oxford University Press. Dumont, M., & Provost, M. A. (1999). Resilience in adolescents: Protective role of social support, coping strategies, self-esteem, and social activities on experience of stress and depression. Journal of Youth and Adolescence, 28, 343–363. DuRant, R. H., Smith, J. A., Kreiter, S. R., & Krowchuk, D. P. (1999). The relationship between early age of onset of initial substance use and engaging in multiple health risk behaviors among young adolescents. Archives of Pediatric and Adolescent Medicine, 153, 286–291. Elliot, D. S. (1994). Serious violent offenders: Onset, developmental course, and termination – The American society of criminology 1993 presidential address. Criminology, 32, 1–21. Elliot, D. S., Hagan, J., & McCord, J. (1998). Youth violence: Children at risk. Washington, DC: American Sociological Association. Fitzpatrick, K. M. (1993). Exposure to violence and presence of depression among low-income AfricanAmerican youth. Journal of Consulting and Clinical Psychology, 61, 528–531. Fitzpatrick, K. M. (1997). Fighting among America’s youth: A risk and protective factors approach. Journal of Health and Social Behavior, 38, 131–148. Fitzpatrick, K. M., & LaGory, M. (2000). Unhealthy places: The ecology of risk in the urban landscape. New York, NY: Routledge. Flisher, A. J., Kramer, R. A., Hoven, C. W., King, R. A., Bird, H. R., Davies, M., Gould, M. S., Greenwald, S., Lahey, B. B., Regier, D. A., Schwab-Stone, M., & Shaffer, D. F. (2000). Risk behavior in a community sample of children and adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 881–887. Garrison, C. Z., Addy, C. L., Jackson, K. L., McKeown, R. E., & Waller, J. L. (1991). The CES-D as a screen for depression and other psychiatric disorders in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 636–641. Gephart, M. (1997). Neighborhoods and communities as contexts for development. In: J. B. Gunn, G. J. Duncan & J. L. Aber (Eds), Neighborhood Poverty, Context and Consequences for Children (Vol. 1, pp. 1–43). New York: Russell Sage. Gibbons, F. X., Larsen, M. H., & Gerrard, M. (1995). Prevalence Estimates and Adolescent Risk Behavior: Cross-cultural Differences in Social Influence. Journal of Applied Psychology, 80, 107–121. Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Beverly Hills, CA: Sage. Graham, J. W., Marks, G., & Hansen, W. B. (1991). Social influence processes affecting adolescent substance use. Journal of Applied Psychology, 76, 291–298. Green, C. A., & Pope, C. R. (2000). Depressive symptoms, health promotion, and health risk behaviors. American Journal of Health Promotion, 15, 29–34. Greenberger, E., Chen, C., Tally, S. R., & Dong, Q. (2000). Family, peer, and individual correlates of depressive symptomatology between U.S. and Chinese adolescents. Journal of Consulting and Clinical Psychology, 68, 209–219. Grob, A., Wearing, A. J., Little, T. D., & Wanner, B. (1996). Adolescents’ well being and perceived control across 14 sociocultural contexts. Journal of Personality and Social Psychology, 71, 785–795.

210

KEVIN M. FITZPATRICK ET AL.

Group for the Advancement of Psychiatry, Committee on Preventive Psychiatry (1999). Violent behavior in children and youth: Preventive intervention from a psychiatric perspective. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 235–241. Hawkins, J. D. (1995). Controlling crime before it happens: Risk-focused prevention. National Institute of Justice Journal, 229, 10–18. Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 12, 64–105. Hawkins, J. D., Herrenkohl, T. I., Farrington, D. P., Brewer, D., Catalano, R. F., Harachi, T. W., & Cothern, L. (2000). Predictors of youth violence. Washington, DC: Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, United States Department of Justice. Henry, B., Avshalom, C., Moffitt, T. E., & Silva, P. A. (1996). Temperamental and familial predictors of violent and non-violent criminal convictions: Age 3 to age 18. Developmental Psychology, 32, 614–623. Holland, A., & Andre, T. (1987). Participation in extracurricular activities in secondary school: What is known, what needs to be known? Review of Educational Research, 57, 437–466. Hui, C. H., & Triandis, H. C. (1989). Effects of culture and response format on extreme response style. Journal of Cross-Cultural Psychology, 20, 296–309. Jessor, R. (1993). Successful adolescent development among youth in high-risk settings. American Psychologist, 48, 117–126. Jessor, R. (Ed.) (1998). New perspectives on adolescent risk behavior. Cambridge, UK: Cambridge University Press. Jessor, R., & Jessor, S. (1977). Problem behavior and psychosocial development. New York, NY: Academic Press. Johnson, P. B., & Johnson, H. L. (1999). Cultural and familial influences that maintain the negative meaning of alcohol. Journal of Studies on Alcohol, 13S, 79–83. Johnson, R. A., & Hoffmann, J. P. (2000). Adolescent cigarette smoking in U.S. racial/ethnic subgroups: Findings from the national education longitudinal study. Journal of Health and Social Behavior, 41, 392–407. Kelley, B. T., Huizinga, D., Thornberry, T. P., & Loeber, R. (1997). Epidemiology of serious violence. Washington, DC: Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, U.S. Department of Justice. Kessler, R., & Neighbors, H. W. (1986). A new perspective on the relationships between race, social class, and psychological distress. Journal of Health and Social Behavior, 27, 107–115. Kovacs, M. (1983). The children’s depression inventory: A self-rated depression scale for school-aged youngsters. Unpublished manuscript. University of Pittsburgh, School of Medicine. Kovacs, M. (1992). Children’s depression inventory (CDI). Toronto: Multi-Health Systems. LeClere, F. B., Rogers, R. G., & Peters, K. (1997). Ethnicity and mortality in the United States: Individual and community correlates. Social Forces, 76, 169–198. Levine, F., & Rosich, K. J. (1996). The social causes of violence: Crafting a science agenda. Washington, DC: American Sociological Association. Loeber, R., Farrington, D. P., Stouthamer-Loeber, M., & Van Kammen, W. B. (1998). Multiple risk factors for multiproblem boys: Co-occurrence of delinquency, substance abuse, attention deficit, conduct problems, physical aggression, covert behavior, depressed mood, and shy/withdrawn behavior. In: R. Jessor (Ed.), New Perspectives on Adolescent Risk Behavior (pp. 90–149). Cambridge, UK: Cambridge University Press.

A Tale of Two Cities

211

Maguin, E., Hawkins, J. D., Catalano, R. F., Hill, D., Abbott, R., & Herrenkohl, T. (1995). Risk factors measured at three ages for violence at age 17–18. Paper Presented at the American Society of Criminology, November, Boston, MA. Masi, G., Tomaiuolo, F., Sbrana, B., Poli, P., Baracchini, G., Pruneti, C. A., Favilla, L., Floriani, C., & Marcheschi, M. (2001). Depressive symptoms and academic self-image in adolescence. Psychopathology, 1, 57–61. Masten, A. S. (1994). Resilience in individual development: Successful adaptation despite risk and adversity. In: M. C. Wang & E. W. Gordon (Eds), Educational Resilience in Inner-city America: Challenges and Prospects (pp. 3–23). Hillsdale, NJ: Lawrence Erlbaum. McArdle, P., Wiegersma, A., Gilvarry, E., McCarthy, S., Fitzgerald, M., Brinkley, A., Kolte, B., Johnson, R., & Quensel, S. (2000). International variations in youth drug use: The role of individual behaviours, peer and family influences and geographical location on drug use in young people. European Addiction Research, 6, 163–169. McCord, J., & Ensminger, M. (1995). Pathways from aggressive childhood to criminality. Paper Presented at the American Society of Criminology, Boston, MA. Petraitis, J., Flay, B. R., Miller, T. Q., Torpy, E. J., & Greiner, B. (1998). Illicit substance use among adolescents: A matrix of prospective predictors. Substance Use and Misuse, 33, 2561–2604. Piko, B. (1998). Social support and health in adolescence: A factor analytical study. British Journal of Health Psychology, 3, 333–344. Piko, B. (2000). Perceived social support from parents and peers: Which is the stronger predictor of adolescent substance use? Substance Use and Misuse, 35, 617–630. Piko, B. (2001). Smoking in adolescence: Do attitudes matter? Addictive Behaviors, 26, 201–217. Piko, B., & Fitzpatrick, K. M. (2001). Does class matter? SES and psychosocial health among Hungarian adolescents. Social Science and Medicine, 53, 817–830. Prescott, C. A., McArdle, J. J., Hishinuma, E. S., Johnson, R. C., Miyamoto, R. H., Andrade, N. N., Edman, J. L., Makini, G. K., Nahulu, L. B., Yuen, N. Y., & Carlton, B. S. (1998). Prediction of major depression and dysthymia from CES-D scores among ethnic minority adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 495–503. Radloff, L. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. Rapid Intelligence (2004). Crime in Hungary. Online at http://www.nationmaster.com/country/ hu/Crime. Reinherz, H. Z., Giaconia, R. M., Hauf, A. M., Wasserman, M. S., & Paradis, A. D. (2000). General and specific childhood risk factors for depression and drug disorders by early adulthood. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 223–231. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Rosenberg, M. (1979). Conceiving the self. New York, NY: Basic Books. ´ Koml´osi, A. V., G´adoros, J., & Csorba, J. (1999). Gyermek e´ s serd¨ul kori depresszi´o R´ozsa, S., Vetr´o, A., k´erd ´ıves m´er´es´enek lehet s´ege a klinikai e´ s normat´ıv mint´an szerzett tapasztalatok alapj´an (The possibility of measuring child and adolescent depression based on experiences gathered on clinical and normative samples.). Pszichol´ogia, 19, 459–482. Sampson, R., & Laub, J. (1993). Crime in the making: Pathways and turning points through life. Cambridge, MA: Harvard University Press. Schneider, B. H. (1998). Cross-cultural comparison as doorkeeper in research on the social and emotional adjustment of children and adolescents. Developmental Psychology, 34, 793–797.

212

KEVIN M. FITZPATRICK ET AL.

Smith, C., & Thornberry, T. P. (1995). The relationship between childhood maltreatment and adolescent involvement in delinquency. Criminology, 57, 451–481. Smucker, M. R., Craighead, W. E., Craighead, L. W., & Green, B. J. (1986). Normative and reliability data for the children’s depression inventory. Journal of Abnormal Child Psychology, 14, 25–39. U.S. Bureau of the Census (1993). 1990 Census of population and housing: Population and housing characteristics for census tracts and block numbering areas, Birmingham, AL MSA. Washington, DC: U.S. Government Printing Office. Warrren, C. W., Karrn, L., Small, M. L., Santelli, J. S., Collins, J. L., & Kalbe, L. J. (1997). Age of initiating selected health-risk behaviors among high school students in the United States. Journal of Adolescent Health, 21, 225–231. Way, N., Stauber, H. Y., Nakkula, M. J., & London, P. (1994). Depression and substance use in two divergent high school cultures: A quantitative and qualitative analysis. Journal of Youth and Adolescence, 23, 331–357. Williams, D. R. (1996). Introduction: Racism and health. A research agenda. Ethnicity and Disease, 2, 126–141.

MAPPING CHANGES IN THE SOCIAL LIVES OF CHILDREN: AN ANALYSIS OF ACTIVITIES AND BEST PRACTICE INITIATED THROUGH SCHOOL-BASED MENTORING IN PHILADELPHIA Suellen Butler INTRODUCTION What kinds of changes do mentoring programs initiate for children and their schools? According to a study by Public/Private Ventures which specializes in social policy, just under half of mentoring programs are based at a particular site (Herrera, 1999). Schools are the most common site recognized. Schools purchase the mentoring program examined in the study to follow. Adults work with children primarily on academic activities and they are paid to do so. The school pays half of the cost of the NSCC (National School and Community Corp) program and federal funds provide the remaining support. Herrera (1999, p. 11) has argued the most direct benefit of school-based programs is the mentor’s integration into the school environment and the role of advocate that the mentor can play. Advocacy has been recognized as the critical task performed by mentors. As advocate, mentors develop a relationship with a child, which advances youth development and deters youth detrimental behavior (Herrera, 1999). This singular relationship, the advocate mentor and child, is challenged by research that follows. Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 213–229 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10011-1

213

214

SUELLEN BUTLER

The advocate mentor recognizes a socialization model used in studying children that assumes the mentor as the active role taker shaping the child as passive recipient of mentoring. This model relies on an adult perspective, assuming the child to be an empty vessel serving as clay to be shaped (Waksler, 1994). Waksler (1994) has criticized this model, arguing that children must be taken seriously as a topic of study – not as objects of socialization. The interpretive model of childhood development by contrast is more useful for examining the data gathered in the following study. Rather than focus on the advocate mentor in relationship with a child, the interpretive model recognizes the child as actor and highlights the range of peer relationships sponsored through mentoring activities. This collective view as contrasted with the singular relationship assumes a different perspective on childhood development. Corsaro’s cross cultural studies (1986, p. 235) of children in preschool settings engage the collective view recognizing childhood development as, “a productivereproductive process of increasing density and reorganization of knowledge that changes with cognitive and language abilities and with changes in their social world.” Rather than highlight the advocacy and one-on-one singular relationship, research reported in this study finds that school-based programs initiate changes to the social world of children supporting involvement expansion and production of peer culture. Research findings reported in this study indicate that the best (mentoring) practices identified by teachers are collective activities that children engage as they conduct negotiations that reconfigure school social life. Mapping this process – how it is engaged by children, expands their social life and transforms their schools – is a study of growing involvement framing the investigation that follows. A series of factors, more common to urban schools, deter children from engaging and expanding involvement. Rather than supporting involvement they foster isolation. Large class size, for example, in combination with sagging achievement levels result in a classroom format that relies on seatwork assignments and working alone. Teachers devote much of their attention to children who are not at basic level and to children who have trouble focusing on task and staying in their seats. According to a recent report on urban schools less than a quarter of urban students can read at basic levels and barely a third can do basic math and science (Saunders, 2000). Establishing the seatwork assignment as the dominant pattern in the urban classroom, results in children working in isolation or in dyadic relationships with the teacher. Beyond achievement levels and structured classroom format, isolation in urban schools is associated with teacher inexperience. There is clear evidence indicating that new, less experienced teachers are disproportionately assigned to urban schools (Saunders, 2000). Large school and class size, the number of children who are not at basic achievement level, seatwork assignment format and inexperienced teachers in

Mapping Changes in the Social Lives of Children

215

combination promote a climate of isolation that counters involvement greatly reducing the social life of children in urban schools. The isolation fostered by this series of factors put children at risk by increasing their potential to invest in an “oppositional culture” described by Anderson (1999). Elementary and middle school teachers interviewed by Anderson (1999) reported increasing investment by children in an “oppositional culture.” According to the teachers interviewed children in the first grade show little investment in an “oppositional culture” that supports the Code of the Street. Yet, by the fourth grade teachers recognize widespread investment by children in the oppositional culture described by Anderson. Anderson’s neighborhood observations and interviews in the schools indicate it is not simply isolation children risk in urban schools but that as they advance in elementary school there is increasing probability of their investment in an oppositional culture that rejects the importance of education in favor of a code protecting children on the street. The code recognizes “toughness” “respect” and the ability to threaten violence as essential tools in negotiating public life (Anderson, 1999).

STUDIES IN CONSTRUCTING INVOLVEMENT Three observational studies serve as a framework for examining the involvement process that children engage and expand as they transform school social life. First, Mandell’s observations (1994) of children in day care centers defines involvement as a continuum ranging from private meaning to publicly held understanding. Children in the day care setting are observed and described as they conduct negotiations that move private, hidden meaning to publicly shared understandings. Mandell’s observations guided by Mead’s conception of “taking account” (1939) yield an identification of four stances of involvement that children assume in shifting hidden private meaning to publicly shared understandings. The first stance, self-involvement, children were observed as isolated. Private meaning prevails at this level as children adapt their speech to the objects (toys and drawings) that they are addressing. “Involvement from afar” characterizes a second stance in which children monitor and follow the ways of others, demonstrating a peripheral commitment. At this level, Mandell (1999, p. 16) describes children as eager to take in as “much information as possible about children and yet not willing to participate with another child.” Comparing it to Becker and Geer’s (1968) notion of “learning the ropes,” Mandell (1999) claims, “children move in and out of this involvement stance constantly and use it as a well defined period of quiet observation and reflection” (p. 169). Staring is characteristic of this stance of involvement. In addition, display, public monologues, and announcements, serve

216

SUELLEN BUTLER

in making initial overtures to others. Mandell (1994, p. 169) notes, there is “a move to cast others in the stance of audience.” Co-involvement distinguishes a third stance. Mandell describes, “digging” as the central process at this stage as children attempt to track public actions of others and fit or join their actions. At the level of co-involvement, children are not able to fully take into account the feelings, attitudes, and interpretations of others. Nor are they prepared to articulate and build on them. However, the move toward public understandings – the foundations of a shared social world – is clearly recognized in the co-involvement stance. It is only at the fourth level, reciprocal involvement, that children, acting jointly, create shared definitions and evidence of a fully defined social world designed by and for them. Moving beyond co-involvement, children reciprocally involved are able to take into account the feelings, attitudes, and interpretations of other children. Although Mandell does not recognize this stance as signaling the production of a peer culture her observations are confirmed through additional studies of children that do (Corsaro, 1997). The different levels of involvement observed in the day care center are used to classify stories children draft and share as they learn to write and use written language in a first grade writing workshop (Butler, 1999). Stories are drafted to share with an audience of workshop writers and consequently the self-involvement stance is practiced during the drafting period but all drafts potentially will be shared with the workshop. As a consequence children draft their first stories addressing others through a perspective that proclaims the “things I know about.” Assuming this stance they begin the process of disclosing private meaning. It is in the personal claim to experience (things I know about) that children begin to relate to others and gain practice with the first person perspective. Stories titled “My Books” and “I Like Stories” clearly indicate the initial stage of the “taking account” process (Mead, 1939). In these accounts beginning writers attempt to relate personal feeling and move toward making private meaning public as they address the workshop audience. Stories from a second round of writing demonstrate a shift to co-involvement. Children use stories to track public actions of others by fitting or joining their actions in a second phase of the “taking account” process. The fitting or joining develops through drafts of the modular story. Remixing story parts through repetition of story problems with slightly different beginnings and endings expands narrative practice. The “lost stories” exemplify the move to act jointly and create shared definitions. Using stories to affirm “things we know about,” children conduct the “dig” for meaning that moves stories in the direction of expanding public understanding. Repeating similar story problems with different characters, beginnings and endings signals participation in a growing field of

Mapping Changes in the Social Lives of Children

217

shared understanding. In a final round of writing children begin to use stories to explore the world. Children using writing to explore the world engage reciprocal involvement and draft narrative that is without precedent. This final round of writing signals the move to achieving public understandings which support a child in drafting experience that is imagined. The end of the story, not known at the start, distinguishes this level of writing. Much of what is drafted in the third round has yet to be affirmed by the workshop. There is risk associated with sharing that which is without precedent, because stories from the third round may introduce ideas that others may not be familiar with and therefore raise questions about. Achieving some level of public understanding based on the joint actions and writing of the second round offers assurance to children who engage risk by drafting a story that is without precedent. Children are able to take into account the feelings, attitudes and interpretations of others as they engage reciprocal involvement and draft narratives whose endings are not known at the start of the stories. Public understandings develop as children request and accept assistance in drafts that break with precedent. Bruner’s (1986, p. 24) research links the involvement observations from the day care center with the narrative from the first grade workshop recognizing and confirming the importance of the rounds of writing and shifting levels of involvement by observing, “It is not just that the child must make his knowledge his own, but that he must make it his own in a community of those who share his sense of belonging to a culture.” Corsaro’s research, recognizing Bruner’s principle of knowledge gained in a community, builds on the involvement continuum observing that the routines practiced by children, much as rounds of writing produce peer culture. Using cross cultural observation of children in both Italian and American preschool, Corsaro identifies the basic patterns of cultural production and reproduction (Corsaro, 1988, 1991, 1997). Through their production of and participation in peer routines children attempt to transform confusions and ambiguities from the adult world into familiar and shared routines of peer culture (Corsaro, 1988). This process of transforming confusions and ambiguities identified by Corsaro corresponds with Mandell’s definition of the shift from privately to publicly shared meaning in the day care center and the process of using workshop stories initially to voice “things I know about” and then using workshop stories to confirm “things we know about” and finally using stories and the workshop setting to explore “things I want and we should find out about.” Moving private meaning to public understanding through peer routines that transform confusion and ambiguity generates growing involvement and the production of peer culture. More recently Corsaro’s discussion of “priming events” provides additional support for the significance of collective action in childhood development. Using an agricultural metaphor Corsaro likens the priming event as “fertile soil in which

218

SUELLEN BUTLER

children’s social representations of coming transitions in their lives takes root and are nurtured through collective action with adults and peers” (Corsaro, 2000). Recognizing that priming events are influenced by educational policies, Corsaro and Molinari (2000, p. 18) defines them as, “activities in which children by their participation attend prospectively to ongoing and anticipated changes in their lives.” Negotiations in the day care center, the cross cultural preschool setting and the writing workshop recognize children as fully human actors engaged in a process that they conduct which yields publicly shared understandings. Together these observational studies confirm the interpretive model (Corsaro, 1991, p. 235) recognizing childhood development as a “productive – reproductive process of increasing density and reorganization of knowledge that changes with a child’s developing cognitive and language abilities and with changes in their social world.” The study that follows moves the analysis further to examine changes in the social world of children – changes introduced through school based mentoring.

DATA COLLECTION Data for the study were gathered through observation and interviews conducted over a six-year period (1995–2000) while serving as an observer of AMERICORP programs in Philadelphia’s public schools. Employed by an external evaluator of AMERICORP’s NSCC (National School and Community Corp) the author conducted both baseline and end of year visits to six urban elementary schools that sponsored the NSCC school-based mentoring program. Conducting an observational visit of a school based mentoring program consists of a series of data gathering measures. The observational visit is arranged and scheduled through an NSCC team leader who directs the mentoring program at each site. A team of five or more corpmembers and a team leader staff each school-based program. Baseline observations of the mentoring program are conducted in October and November and consist of both fixed and flexible components. The fixed component is a set of interviews with team leader, principal, teachers, and corpmembers serving as mentors. There are specific questions in the interview asked of teachers and principals seeking description and identification of best practices as well as activities deemed a “waste of time.” The flexible component of the baseline includes brief talks with teachers and observations of best practices and program activities inside as well as outside the classroom. End-of-year observations conducted in May and June consist of questions directed at evaluating the effectiveness of the program. Whereas the baseline interviews seek to determine the extent of program and activity implementation the end-of-year observations and interviews focus on

Mapping Changes in the Social Lives of Children

219

effectiveness and satisfaction with the established program and activities. All program participants are assured of the confidentiality of the information gathered during the evaluation process. No names are cited in any part of the data collection. The interviews and observations are distinguished by role (team leader, principal, teacher, corpmember or mentor) and mentored activity. An observational day at a school based mentoring site begins with an interview with the mentoring program team leader. This interview provides an introduction to the range of activities designed and implemented by the mentoring team. Following the team leader interview the school principal interview is conducted. Once the principal has been interviewed the remainder of the day is devoted to interviewing teachers, asking specifics about the mentoring program activities they are familiar with and interviewing members of the mentoring team. In addition to these fixed interviews, observations of mentored activities are conducted during and after school.

Analysis of Mentoring Activities Three activities are common to mentoring programs observed over the six-year period: tutoring programs, socialized recess and after school clubs. Examining these activities permits addressing the larger question – What kinds of changes do mentoring programs initiate for children and their schools? The tutoring activities conducted by mentors vary across school programs. Some children identified by teachers and resource staff, are tutored by mentors in the NSCC office space. This is common where office space is expansive including chalkboard and furniture that accommodate tutoring. At other schools, children are tutored in the library as individuals or in small groups. Some NSCC mentors tutor in the classroom working with children that request help during problem solving sessions. In one school where space was a major problem, a tutoring table was set up in the second floor hallway. Two mentors conducted drill and discussion sessions with small groups of second graders. In newer buildings, pod areas outside a cluster of classrooms provide tutoring space as well as wall space to display charts recording a child’s attendance and progress at tutoring sessions. Workshop-like conditions and the “digging” that engages negotiation and shifts in involvement are more likely to transpire in the small group tutoring sessions conducted by mentors outside the classroom. It is here that children find greater flexibility for their negotiations with other children. The small groups of 4–6 children tutored by two NSCC mentors in the second floor hallway or the library or NSCC office permit children the opportunity to track public action and fit or join actions with others in a search for common

220

SUELLEN BUTLER

ground. Children are granted more freedom to discuss assignments as a group, share answers and reflect jointly on solutions. The bumpy and disjointed efforts generated by these actions are indicative of the co-involvement stance described by Mandell (1994). The co-involvement stance and the negotiations between children permitted in mentored tutoring sessions are uncommon in the classroom. There was a definite preference for seatwork assignments found in most classrooms entered during observational visits. The pattern is insured by two factors. First, classes are large, often with 33–35 children to a classroom. Beyond the obstacle of large class size, is the problem of wide ranging achievement levels. Many of the children in schools that sponsor NSCC programs are performing, according to standardized test scores, below achievement levels. One principal recognized the importance of NSCC tutoring and support for nearly 70% of her fourth graders who tested below basic reading level. Teachers devote much of their attention to children who are not at basic level and to children who have trouble focusing on tasks and staying in their seats. Established as the dominant pattern inside urban classrooms, seatwork assignments require that children work independently. It is nearly impossible to conduct negotiations and search for common ground when the classroom is structured so children work alone. Restrictions imposed on negotiations between children in the classroom yield “involvement from afar.” Negotiations that do occur in the classroom are almost exclusively dyadic between teacher and individual child. Learning in a classroom where nearly all interaction is dyadic and where seatwork assignments are common practice calls for little in the way of negotiation between children and makes little headway in moving private hidden meaning in the direction of publicly shared understanding a process essential to production of peer culture. Socialized recess, a second activity conducted across all school sites, offers children the basic component of co-involvement or the opportunity to search for common ground. Mentors organize games, maintain boundaries for designated age groups, mediate problems or arguments that arise maintain and distribute recess equipment, line-up and direct children in and out of school buildings. Some but not all NSCC mentors assist in the lunchroom helping to seat children and facilitate in the systematic distribution of lunches, lunchroom clean-up and conduct contests with weekly parties and prizes. Principals recognized socialized recess as an “essential” activity, necessary for “reducing the structure of the classroom – but reintroducing it at the end of recess.” Mentors get to know every child in the school through socialized recess and lunchroom programs. This setting permits mentors to work with small groups and through games in the schoolyard and gym as well as contests in the lunchroom develop activities that support negotiation and permit shifts in the level of group involvement. Two of the school principals1 reported

Mapping Changes in the Social Lives of Children

221

socialized recess as the best practice designed by corpmembers arguing that orderly play time after lunch prepared children to come back to the classroom prepared to learn without such assistance one of the principals reasoned, “We loose a day of learning.”2 Mentors cited the socialized recess program as essential for “cutting down on bullies and fighting while promoting teamwork, sharing and competition (Wells, 2001).” One teacher recognizing the merits of socialized recess described the activity as a place where “children learn how to get along.”3 One incident observed outside of the schoolyard and lunchroom is indicative of co-involvement generated in these settings. While waiting in the front office to interview a principal one morning, I sat with a nine-year-old boy who had been sent to the Principal’s office. He whispered threats and insults to the boys and girls who brought class attendance sheets to the office. When I asked him why he was sent to see the Principal, he said, “Be quiet.” And when an office administrator brought him lunch from the cafeteria he yelled, “I don’t want it.” When the team leader of the mentoring program came to check on me and the day’s itinerary, the boy ran to him calling his name, holding his arm and asking, “Can we go to the gym today?” The excitement and enthusiasm that he voiced were so different from all that I had observed and for a moment I caught a glimpse of the kind of joy and delight indicative of an involvement stance that reflects common ground and publicly shared understanding. The team leader put his arm around the boy’s shoulders and said, “We’ll have to check the schedule Anthony – see you later.”4 Anthony returned to his seat, unwrapped and started to eat the cookie that was part of his lunch. The Principal came out to get me for the interview and as I walked toward her office I looked at Anthony spearing his bag of chocolate milk and said, “Bye, Anthony.” He said nothing, making it clear there was no change in his manner toward me. That brief interaction with the team leader was the only positive expression that Anthony made in the 40 minutes we spent together in the front office. Yet, I heard him use the word “we” affirming his relationship in at least one setting where Anthony had negotiated common ground and practiced a level of co-involvement that moves private meaning to publicly shared understandings signaling production of peer culture. After school clubs are the third activity sponsored by mentors across school sites. The Homework Club, Friday Funday, the Student Yearbook, the hospital service learning project, the talent and fashion show, Drama Club, Books and Beyond, Sister-Sister, Girl and Boy Scouts and the Drill, Basketball and Soccer teams are a few of the after school activities implemented by the mentoring program. These activities unlike those during the school day are often well photographed by mentors. Bulletin boards inside and outside of the mentoring office often exhibit photographs of recent activities. Photographs examined across school sites provide an indication of the range of settings and the array of negotiations that children

222

SUELLEN BUTLER

conduct in the after school activities sponsored by the mentoring program. Many settings are not immediately identifiable in the photographs, but there are clues conveyed through lighting, clothing, and d´ecor that speak directly to the range of sponsored activities. It is immediately apparent that these are not photographs of seatwork assignments and the restricted activities imposing “involvement from afar.” Most instructive are the action shots since they are photographs where the intent has been to record group activity and not individuals. Here is evidence that documents children in the midst of negotiations and co-involvement. The action shots capture groups of children in the “digging” process characterized as the bumpy and disjointed efforts to join actions with others in search for common ground. The group activity photographs generate the greatest response from children passing by the bulletin board. It is not so much that they stop to study the activity but to affirm the children they know and recognize in the photographs. It is difficult to do, since the photographed children are not looking to the camera, presenting themselves as individuals, but rather caught in the involvement construction process, engaged by joint actions that reflect the process of shifting private meaning to publicly shared understandings. The photographic records posted by the mentoring program document the opportunities for negotiation at the same time reflecting the common ground children establish through activities sponsored by NSCC mentoring programs.

ENGAGING AND EXPANDING INVOLVEMENT School based mentoring programs introduce two critical elements that facilitate growing levels of involvement that move private meaning to publicly shared understanding and reduce social isolation in urban schools. First, mentoring programs introduce a distinct set of roles and role relationships. The team leader and mentors are clearly distinguishable from schoolteachers, school administrators and NTA’s (non-teaching assistants). The mentor role has been characterized in interviews with teachers and principals as “more of a friend.”5 Mentors are addressed by their first names or in some schools by the addition of Miss or Mr. to first names. Members of the mentoring program are neither addressed nor perceived in the formal manner that teachers are. When I asked one teacher about the role she told me, “I can’t do what she is doing – I cannot afford to be causal.” “I wish that I could but I cannot.”6 An adult as friend assumes a more helpful role and for this reason does not claim or impose the same difference in authority as the teacher in relationships with children. Mentors neither assign grades nor report STAT-9 scores. They are

Mapping Changes in the Social Lives of Children

223

adults, yet they do not assume the authority that teachers and administrators do. Rather, they establish a more equitable exchange that facilitates negotiations and co-involvement between children. Unlike NTA’s (non-teaching assistants), mentors are not viewed by children as security personnel. Observations in one school provide a telling incident reported by a principal. Teachers enlisted members of the mentoring program to staff the detention center. The team leader declined the offer to serve in the detention center explaining the intent of the mentoring program would be jeopardized were mentors to assume the role of detention center supervisor. The team leader, it was explained viewed the proposed role expansion as a threat to the mentoring relationship requiring a shift in authority and additional punitive expectations. Describing the role and status one mentor explained during an interview, “I don’t yell but I ask questions – why are or were you acting like this?” “But after focusing on their inappropriate behavior and reminding them that they are the losers if they give up their education – I build them back up.” “They have to know that I wouldn’t put up with their misbehavior –but that I am a friend.” This same mentor explained that, “many of these children have little or no structure at home – then they come here and all they get is yelling, rules and disrespect.”7 School-based mentoring programs offer children a relationship that is without precedent in their lives. Neither teacher limited by the formal constraint of the classroom, nor working poor parent, struggling with the economic constraint of the household, can offer their children this mix of critical acceptance. Described by the interviewed mentor, critical acceptance is a process of recognizing misbehavior, calling for reflection, yet remaining steadfast in support of a child. Mentoring programs introduce a second element that engages involvement and reduces social isolation inside schools. Mentors design and define group settings that would otherwise not be part of the school social life. Tutoring groups, socialized recess groups, after school clubs, each group setting designed to permit children room to negotiate with other children and search for common ground that transforms private meaning to publicly shared understandings. Schoolbased mentoring programs specialize in designing and defining group settings and role relationships that provide children with the opportunity for “digging” and the search for common ground that generates a climate of involvement. Mentoring programs provide a setting for the negotiation of experience. What does the negotiation of experience sound like? It sounds “bumpy and disjointed” in the words of Mandell like observations recorded at an after school meeting of the Sister Sister Club where girls from grades 3 to 5 come, “to talk, make things, and go places – together.”8 It is that added dimension of “together” recognized by the girl interviewed at the club meeting that supports construction of involvement.

224

SUELLEN BUTLER

Two activities cited as “best practices” in interviews with teacher recognize mentoring activities that typify growing involvement and the production of peer culture. Mrs. J, an experienced teacher with more than 20 years of teaching at the school site, recognized the “Justice and Kindness project” conducted by a mentor in her fourth grade class. She noted, during the end of year interview, the absence throughout the school year of the long-term (several weeks) small group project. Ten children selected by Mrs. J, worked with the mentoring team leader conducting a study of advertising and advertisements and then set about to conduct their own advertising campaign – designing and recording public service ads to promote justice and kindness. When I asked Mrs. J why this project qualified as a best practice, she told me, “Involvement in the entire process – research, writing, explaining, advertising, recording, showing results to other children throughout the school – involvement over several weeks leading to culmination.”9 Mrs. J describes the essential ingredient as involvement – “involvement in the entire process” and again “involvement over several weeks leading to culmination.” “This is what they do best, she told me – sponsor involvement.” It is clear that working for several weeks on an advertising campaign for justice and kindness provides ten children the opportunity for the construction of involvement. Yet these 10 youth are a very small portion of the nearly 800 children attending the elementary school. The children from the “Justice and Kindness Project” took their advertising campaign to children in first and third grade classrooms, explaining what they did, then showing and explaining their work through video ads for Justice and Kindness. This effort brings small group work to large classrooms and moves the publicly shared understanding through an expansion process constructed in the small group introduced to the larger groups, the group of 10 acting as “purveyors of peer culture.” Best practices move beyond co-involvement as children prepared to take into account the feelings, attitudes and interpretations of others, introduce their shared understanding to groups of children throughout the school. Linking the small group to larger groups through publicly shared understandings facilitates a transition in school social life as children engaging other children, purvey bits of peer culture produced through mentored activities. These efforts support reciprocal involvement an expansion described by Mandell (1994) in observations of the day care center. The best practices of school based mentoring afford children opportunities for active role taking – taking some of what they have learned, some of what they are good at and putting it to the test conducted by introducing it to larger groups of children. Again, Bruner’s (1986, p. 24) observation that, “it is not just that the child must make his knowledge his own, but that he must make it his own in a community of those who share his sense of belonging to a culture.” It is through such testing of publicly shared understandings that children

Mapping Changes in the Social Lives of Children

225

conduct an expansion process at two levels. First, at the individual level, they expand their practice as active role takers challenged by opportunities to take on other perspectives, shifting back and forth engaging multiple perspectives. And second, they deposit the shared understanding derived through this process within a larger cycle regulating production and reproduction of peer culture. Expanding and shifting involvement children move toward publicly shared understanding engaging a process that transforms school social life. The second best practice illustrating involvement expansion and production of peer culture sponsored through mentoring activities was described during an interview with a reading resource teacher.10 Mentors specializing in literacy receive training for implementing the 100-Book Challenge reading program. Two mentors specializing in literacy introduced the 100-Book Challenge during the summer camp program conducted at the elementary school throughout the month of July. The following year the reading resource teacher, recognizing the programs benefit for the summer camp youth, submitted a small grant to expand the 100-Book Challenge to four of the first grade programs. The mentors initiated and conducted training for two classes of third grade children. These children were trained as reading coaches for the first grade classes. The training consisted of showing the older children how to listen, help and record information about the books that their first grade partners read over the course of the semester. The third grade teacher was very enthusiastic about the program recognizing the benefit for her third graders especially those children who were not quite at third grade reading level. Their interest and enthusiasm for reading was spurred along by their role as reading coach for the new readers from the first grade. “It’s the confidence factor,” the teacher noted.11 The benefits for the first graders, partnered with the “big kids” from third grade again demonstrate the reconfiguration of the classroom into partner groups where co-involvement and the “dig” for meaning shifts private meaning toward publicly shared understanding. The reading coach’s best practice recognizes a setting designed through mentoring that permits involvement expansion and increased potential for negotiating shared understandings. Best practices recommended by interviewed teachers describe “children as active role takers.” Conducting the research on advertising and developing an advertising campaign on justice and kindness, the ten children from Mrs. J’s class, experience the shift in involvement levels, moving private meaning toward the publicly shared understanding that yields common ground. Established in the small group, common ground yields publicly shared definitions that are engaged and spread through larger groups. The process corresponds to Corsaro’s (1997) description of the production and reproduction of peer culture. The advertising experts present their campaign for justice and kindness in classrooms throughout the school. In much the same way, third grade reading coaches negotiate

226

SUELLEN BUTLER

co-involvement as they take into account the feelings, attitudes and interpretations of their partners prepared to build on them in settings sponsored through school based mentoring. The best practice of the school based mentoring program functions as a priming event, “activities in which children by their participation attend prospectively to ongoing and anticipated changes in their lives (Corsaro & Molinari, 2000, p. 19).” The reading coaches and their beginning reader partners – the child from Honduras in the adopt-a-grandparent program, the justice and kindness campaign workers – children in each mentored activity reflect the importance of priming events. Corsaro and Molinari (2000, p. 20) observes that, “priming makes the new or strange familiar . . .. Daily meeting times, discussions and several long term projects.” These, Corsaro, Molinari and Rossier (2002, p. 325) observes, are the “subtle ways children were made aware of onging and daily changes in their lives.” The best practice of the school based mentoring program engages children through negotiations and expands involvement beyond the small group serving as a priming event that supports the transitions of childhood development.

CONCLUSIONS Mentors, during interviews, often find it difficult to define mentoring precisely. Some distinguish the activity in the way that they relate to children – as an adult rather than a teacher, suggesting additional agency that children claim. Additional agency like that gained as reading coach and “justice and kindness” campaign manager that builds on co-involvement spreading the construction process beyond the small group setting. Other mentors focus on their efforts to get children to reflect, to see things from other perspectives and to stop, think and raise questions that move private hidden meaning toward publicly shared understanding. “It’s more a mental thing,” one mentor told me. “With tutoring you’re working on skills, but with mentoring you’re working with the whole child.”12 “I have seen a difference,” one teacher explained of her student – referring to a girl from Honduras who “took a lot of teasing from children until she began to attend mentored activities . . .. She got involved in the after school ‘adopt a grandparent program’ at the senior center sponsored by two mentors, and her confidence has grown . . .. She has a stronger sense of herself – these activities give children a chance to find out who they are and what they are good at,” she explained.13 Finding out what you are good at should be the right of every child regardless of the school they attend, the achievement levels of their classmates and the professional experience of their teacher. Finding out who you are through what you are good at is more readily available to children participating in school based mentoring programs. Mentors’ and teachers’ comments convey the program’s

Mapping Changes in the Social Lives of Children

227

promise for engaging and expanding involvement by recognizing children as more than objects of socialization and skill recipients. There is acknowledgement here that school-based mentoring activities attend to a child’s perspective, to their reflection, to their inner discourse. They attend to the active role taker, an attention essential in generating involvement and transforming school social life. Viewing children as active role-takers links the school based mentoring program to the theory and research findings from the daycare center (Mandell, 1994), crosscultural pre-schools (Corsaro, 1988, 1991; Corsaro & Molinari, 2000; Corsaro, Molinari & Rossier, 2002) and writing workshop (Butler, 1999). Observing shifts from individual reflection and private meaning to public understanding builds on co-involvement preparing children to spread involvement beyond the small group through a process that promises peer culture production and reconfiguration of school social life. Opportunities for active role taking in urban schools are expanded when mentors redesign the large classroom, lunchroom or schoolyard into smaller group settings where children granted greater agency overcome isolation as they negotiate the search for common ground. Reconfiguration of social life in urban schools through mentoring programs helps to counter some of the barriers to active role taking generated by large class size, structured classroom format and inexperienced teachers. Granted few opportunities to enter settings and conduct negotiations that allow shifting back and forth between things I know and things we know and from sorting out and internalizing moments that yield both common ground and a developing sense of self, children are barred from actively shifting levels of involvement that combat isolation and counter investment in an “oppositional culture” (Anderson, 1999). Restrictions placed on negotiation of common ground and opportunities for selfevaluation have implications for the social life of children inside urban schools. If children are to engage involvement and combat isolation they need activities that yield settings that grant additional agency and provide opportunities for small group negotiations. They need school based mentoring to assist them in finding out what they are good at. They need the role relationships and settings designed through mentoring for conducting negotiations and declaring publicly some of what they know some of what they share and some of what they want to find out. Sara Mosle (2000, p. 12), a former New York City School teacher and more recently a mentor for a group of former students, criticizes mentoring claiming, “It doesn’t offer a systematic solution to entrenched problems like hunger, poverty, homelessness . . . nor is it a solution for rescuing urban education.” School-based mentoring programs are neither a “cure all” for poverty nor a rescue for urban education. What a mentoring program can do is never clearly recognized by Mosle or by those viewing the mentor as advocate (Herrera, 1999;

228

SUELLEN BUTLER

Sipe, 1998). It holds the potential to reconfigure the social life of children inside urban schools. This is the greatest promise of the school based mentoring program, changing children’s social lives through the design of settings for small groups that generate roles with additional agency and spur shifting levels of involvement that dismantle social isolation.

NOTES 1. Principal M (1998). 2. Principal D (1997). 3. Teacher R (1998). 4. Principle R (1998). 5. Principal S (2000). 6. Teacher T (1999). 7. Mentor K (1999). 8. Student S (1998). 9. Teacher J (1998). 10. Reading Resource Teacher A (2000). 11. Third grade Teacher R (2000). 12. Mentor D (1999). 13. Teacher F (2000).

REFERENCES Anderson, E. (1999). Code of the street. New York, NY: W. W. Norton. Becker, H. S., & Geer, B. (1968). Learning the ropes: Situational learning in four occupational training programs. In: J. Deutscher (Ed.), Among the People: Encounters with the Poor (pp. 208–213). New York: Basic Books. Bruner, J. S. (1986). Actual minds, possible worlds. Cambridge, MA: Harvard University Press. Butler, S. (1999). Active agency and the classroom. In: C. Sheehan (Ed.), Contemporary Perspective Family Research (Vol. 1, pp. 133–156). Stanford, CT: JAI Press. Corsaro, W. (1988). Routines in the peer culture of American and Italian nursery school children. Sociology of Education, 61, 1–14. Corsaro, W. (1991). Peer play and socialization in two cultures. In: B. Scales, M. Almy, A. Nicolopoulou & S. Ervin-Tripp (Eds), Play and the Social Context of Development in Early Care and Education. Corsaro, W. (1997). The sociology of childhood. Thousand Oaks, CA: Pine Forge Press. Corsaro, W., Molinari, K., & Rossier, K. (2002). Zena and Carlotta: Transition narratives and early education in the United States and Italy. Human Development, 45(5), 323–349. Corsaro, W., & Molinari, L. (2000). Priming events and Italian children’s transition from pre-school to elementary school: Representations and action. Social Psychology Quarterly, 63(1), 16–34. Herrera, C. (1999). Relationship development in community based and school based programs. Philadelphia: Public and Private Ventures.

Mapping Changes in the Social Lives of Children

229

Mandell, N. (1994). Children’s negotiation of meaning. In: F. C. Waksler (Ed.), Studying the Social Worlds of Children (pp. 161–178). Philadelphia: Falmer Press. Mead, G. H. (1939). The philosophy of the act. Chicago, IL: University of Chicago Press. Mosle, S. (2000). Mentors introduce to new worlds. New York Times Magazine (July 2, 2000), 22–24. Saunders, W. (2000). The teacher factor. New York Times (Education Supplement, Jan. 9), 12–14. Sipe, C. L. (1998). Mentoring: A synthesis of research 1988–1995. Philadelphia: Public and Private Ventures Publication. Waksler, F. (1994). Studying the social worlds of children. Philadelphia, PA: Falmer Press. Wells, M. (2001). WHYY interview with national school and community corp mentors. October 20.

THE IMPACT OF WAR, ADULT HIV/AIDS, AND MILITARIZATION ON YOUNG CHILDREN’S MORTALITY Steve Carlton-Ford INTRODUCTION The study of war has generally been neglected in sociology, with much of the discussion focusing around military spending or the organization of the military rather than war per se. Sociologists have critiqued and investigated the militaryindustrial complex (Mills, 1959), investigated morale in military units (Durkheim, 1951; Stouffer & DeVinney, 1955), and studied the socialization of soldiers (Cockerham & Cohen, 1980). However, the direct examination of war has been relatively rare. When war has been examined, sociological research has focused on the causes of war, often discussing the preconditions of revolutions (Goldstone, Gurr & Moshiri, 1991; Skopol, 1979), or the reasons for military interventions by core countries in the peripheral countries of the world system (Kowalewski, 1991). Examinations of the sociological impact of war on civilian populations have been even rarer. The work of a few historical sociologists and demographers stands as an exception. For example, Sorokin’s analyses show, across roughly 20 centuries, the fluctuations in the civilian death toll resulting from wars (1941, p. 214). He also details the effects of the Russian revolution, arguing for pervasive negative effects across a broad range of outcomes, including mortality in the civilian population (1941, 1942). Other research, in individual countries, finds that war increases Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 231–255 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10012-3

231

232

STEVE CARLTON-FORD

civilian mortality (e.g. Anderson & Silver, 1989; Merli, 2000), particularly for children (Mielke & Pitkanen, 1989). As Sorokin opined (1941), the twentieth century turned out to be the bloodiest century ever (see Eckhardt, 1990 for both totals and per capita measures), with roughly three quarters of all war-related deaths occurring in the 1900s (Eckhardt, 1991). Previous research suggests that the impact of war on civilian populations has increased and that children are particularly vulnerable (Levy & Sidel, 1997). But this research is either highly aggregated (e.g. by continent) or focused on individual countries (e.g. Vietnam). In contrast, we examine the impact of war on young children’s mortality rates in a large contemporary sample of countries. In the following sections, we discuss the changing composition of war time casualties, focusing on young children. We then examine the relationship of children’s mortality to the potentially confounding or intervening factors discussed below. After discussing our sample, measures, and methods, we estimate the impact of war on young children’s mortality rates while controlling for these other factors. We do not explicitly examine all the causal pathways. However, by examining changes in war’s effects as we control for potential sources of spuriousness, as well as mediating and moderating influences, we better understand how war affects young children’s mortality rates.

WAR AND YOUNG CHILDREN’S MORTALITY Prior to the 1900s, war made relatively little overall impact on death rates for civilian populations. Although there were exceptions, the civilian death toll from wars prior to World War I was usually low in comparison to the wars that followed (Clemens & Singer, 2000). Increasingly during the 20th century, civilians became the casualties of war. By the end of the 1980s, they comprised nearly 90% of all war-related deaths (Bellamy, 1996). These deaths occurred not just among adults, but also among non-combatant children. In civil wars, the majority of wars today, children are often directly targeted. During the late 1980s and early 1990s, approximately 2 million children died as the direct result of armed conflicts (Bellamy, 1996). War also affects children indirectly by destroying public health infrastructures, as well as other social arrangements that might otherwise protect them. Contemporary warfare targets basics, such as sanitation and water, rendering children increasingly susceptible to diseases, many of which are otherwise easily prevented. War also puts children indirectly at risk by killing adults (mothers and fathers), often through the spread of disease.

The Impact of War, Adult HIV/AIDS, and Militarization

233

In addition as is discussed briefly here and in more depth below, economic, social, and cultural contexts can significantly condition war’s effects on children’s mortality. Countries at war in the late 1980s and early 1990s had generally lower levels of economic development, higher levels of military participation (maintenance of troops), and somewhat distinctive religious cultures. Also, these countries have less educated populations, have less developed public health infrastructures, and may have higher adult disease rates than countries without wars. All of these factors indirectly threaten children’s survival, increasing the child mortality rate. Also, each of these factors is related to armed conflict. As a result, any analysis of the impact of war on children’s mortality rates needs to consider these factors simultaneously. Yet, no previous research has done so.

GENERAL ECONOMIC DEVELOPMENT A country’s general level of economic development is probably the strongest single predictor of children’s mortality (Pritchett & Summers, 1996), and is also associated with being at war. Countries at war have lower average levels of economic development (Carlton-Ford, Hamill & Houston, 2000); and shortterm increases in the number of years at war are associated with decreases in economic development (Carlton-Ford, 1999; Collier, 1999). Analyses of several specific conflicts suggest that such economic declines occur because: (1) natural resources are not developed (e.g. for East Timor see Robinson, 1994); (2) war cripples important industries and discourages foreign investment (e.g. for Kashmir see Bhinda, 1994; for the Balkans see Burke & MacDonald, 1994); and (3) mobile forms of capital leave war-torn countries (Collier, 1999). Given the generally higher child mortality rates associated with lower levels of economic development, war-torn countries can be expected to have higher child mortality rates as a result of war’s adverse impact on economic activity. However, the overall effect of general economic development on children’s mortality is not entirely direct; national wealth is translated into lower children’s mortality through a variety of pathways (Pritchett & Summers, 1996), each of which has previously been shown to be affected by armed conflict. The following pathways appear particularly important: high levels of adult HIV/AIDS, increased military participation, promotion of women’s literacy, and development of public health infrastructures. In addition, all of these appear to be related to the religious composition (e.g. percent Muslim) of the countries involved, the final factor considered below.

234

STEVE CARLTON-FORD

ADULT DISEASE-HIV/AIDS In developing countries, where 90% of all HIV/AIDS related deaths occur, HIV/AIDS is spread primarily by heterosexual contact. Infection rates for adult males and adult females in these countries are about equal (Bellamy, 2000, p. 32). Given the high infection rates in some countries, children may either be orphaned by HIV/AIDS or be born infected (Bellamy, 2000, pp. 32–35). AIDS orphans and children with AIDS have higher than average mortality rates. As a result, high rates of adult HIV/AIDS infection in developing nations appear to result in higher levels of child mortality. Rather than being only a recent problem, pediatric AIDS had been identified as a serious health problem in Africa as early as the late 1980s (Preble, 1990). For example, during the late 1980s roughly 10% of AIDS cases in Uganda were among infants and children under the age of five (Gould, 1993). Roughly one quarter to one third of children born to HIV infected women in Africa were expected to die of AIDS (Packard & Epstein, 1992). In addition, HIV/AIDS weakens the immune system making children even more susceptible to death from infectious diseases, like diarrhea and measles, than in wealthier countries. Given the generally low level of health care in many developing nations and the extreme expense of treating AIDS, many children become AIDS orphans when their parents die (De Zalduondo, Msamanga & Chen, 1989; Gould, 1993; UNAIDS, 1996). Currently, AIDS has orphaned over 11 million children under the age of 15 in sub-Saharan Africa; these orphans constitute roughly 80% of AIDS orphans worldwide (UNAIDS, 2003). Children with HIV/AIDS or whose parents have died from AIDS are often the targets of stigmatization, neglect and abuse (UNAIDS, 2002). These orphans often do not receive care and protection from adults because of the medically unfounded fear that they will transmit AIDS to other family members. When healthy AIDS orphans are cared for, often by relatives, they place an added burden on families that are often already economically impoverished (UNAIDS, 1996, 2002, 2003). As a result, children’s mortality rates will probably be higher in countries with high adult HIV/AIDS rates. The countries that bear the heaviest burden of HIV/AIDS lie predominantly in sub-Saharan Africa, an area with a predominance of major armed conflicts. So, controlling for adult HIV/AIDS should partially explain the effects of war on children’s mortality.

MILITARY PARTICIPATION Militarization may drain resources that would otherwise be allocated to health and education (Deger, 1985; Harris, 1997; Levy & Sidel, 1997) or it may adversely

The Impact of War, Adult HIV/AIDS, and Militarization

235

affect well-being more directly (Adeola, 1996). Yet, at least one form of militarization-military participation-appears to reduce young children’s mortality. Some previous research has found that countries with high levels of military participation, that is countries that maintain higher proportions of their population as troops, have lower levels of infant and young child mortality (Bullock & Firebaugh, 1990; Kick et al., 1990). Although the processes have not been clearly specified, maintaining troops probably requires infrastructures (e.g. for food and water, education) that may indirectly help lower children’s mortality rates (Bullock & Firebaugh, 1990; Kick et al., 1990; Scanlan & Jenkins, 2001). In short, this major form of militarization may actually reduce children’s mortality rates. However, the effects of military participation are probably not always positive. For example, transportation networks allow the quick movement of troops, the majority of whom are sexually mature males. Troops have relatively high rates of contact with prostitutes, who have high rates of HIV infection. And in general, troops have higher rates of HIV infection than do civilians in their countries. During wars, conquering troops may spread HIV to defeated populations by rape. As a result, war and military participation are suspected of being transmission vectors for HIV/AIDS (see Foreman, 2002; UNICEF, 2002). The evidence is not, however, unequivocal. In countries with low levels of HIV infection prior to the onset of conflict (e.g. Nicaragua), war appears not to spread HIV/AIDS (Low et al., 1993). If as we argued above, adult HIV/AIDS affects children’s mortality and if military participation is important in the spread of HIV/AIDS only once the rate of infection is relatively high, then we should find detrimental effects of these factors on children’s mortality primarily in countries with both high levels of troops and high HIV/AIDS rates. Given these relationships, any examination of the effects of war on children’s mortality must account for military participation. Whether controlling for differences in infrastructure and HIV/AIDS rates will eliminate the putative advantages of military participation remains to be seen.

WOMEN’S LITERACY Women are, almost universally in the developing world, the caretakers of children. Educated women appear to be more successful in making sure that their children receive better care and stay healthier (Caldwell, 1986, 1990). So it makes sense that women’s literacy strongly predicts children’s mortality (Pritchett & Summers, 1996). Educated women are able to live in neighborhoods that provide better public health through safe water and sanitary conditions (Desai & Alva, 1998).

236

STEVE CARLTON-FORD

Also, it appears that educated women are better able actively to pursue effective medical interventions for sick children (Caldwell, 1996). Regardless of the exact mechanism operating here, having a literate female population is essential to lowering children’s mortality rates. Clearly, however, one target of today’s armed conflicts is the educational infrastructure. School buildings are destroyed and educators are specifically targeted (Zwi, 1996). The longer a country has been at war the less likely it is that its population will be literate. But again, there may be offsetting influences of preparation for war in the form of military participation. Controlling for women’s literacy should partially eliminate any reduction in children’s mortality that accrues from military participation. To the extent that war’s effect on children’s mortality is mediated by women’s literacy, the direct effect of war should diminish when women’s literacy is controlled.

INFRASTRUCTURE In order for children (and others) to remain healthy, they must have safe water and sanitation. The availability of safe water and adequate sanitation (as well as other health infrastructures) depends in part on whether and how long a country has been at war, the general level of economic development, and the presence of a social and cultural climate that promotes health through the development of these infrastructures. As argued above, countries with weaker economies generally have less developed basic public health infrastructures. Cultural factors associated with broad religious traditions may also influence the development of public health infrastructures. War places an extra, often devastating, burden on basic infrastructures. In fact, infrastructures are prime military targets (see Zwi, 1996). During the 20th century, wars have increasingly been fought by destroying the enemy’s ability both to maintain troops and to maintain civilian support for the war effort, rather than through the direct defeat of armies. One of the most effective ways of defeating an enemy has become the destruction of water supplies, sanitary facilities, and health services. This end can be accomplished directly by contaminating water supplies, or in more developed areas, by destroying water purification plants, water pumping stations, and electric generating plants. Both troops and civilians can be cut off from water, sanitation, and medical care by destroying bridges and highways (Zwi, 1996). Health workers often are targeted. Declining economic activity makes purchase of food, water, and medical care (when available) more difficult. Thus, over and above the direct effects of combat, war destroys the public health infrastructure of societies. The reduced public health infrastructure leads to higher mortality rates among children. Controlling

The Impact of War, Adult HIV/AIDS, and Militarization

237

for differences in the public health infrastructures of countries should partially explain war’s effects on children’s mortality.

RELIGIOUS COMPOSITION Broad differences among the four largest religious faiths (Buddhism, Christianity, Hinduism, and Islam) appear to be associated both with the complex of structural variables described above that affect children’s mortality, as well as with war itself. Religious traditions are related to economic well-being, and, as a result, may differentially emphasize traditional norms (Inglehart, 1997) that influence the availability and accessibility of education and public health infrastructures. Many, but not all, of these differences suggest that Islamic countries are, at least at the present, likely to have higher rates of child mortality than other countries with the same general level of economic development. First, predominantly Islamic countries have, in recent decades, been more likely to have a war on their territory (see Abu-Amr, 1997). These conflicts are often the result of economic conditions (Collier & Hoeffler, 1998), with religious differences providing rhetoric for recruiting members (Bowen, 1996; Gurr & Moore, 1997). Where religion appears as a factor in major armed conflicts, the majority involve intra-religious disputes, primarily between rival Islamic groups (Said, 1997). Second, despite early advances in medicine, predominantly Muslim countries currently have less well-developed public health infrastructures (Caldwell, 1996). In general, better developed public health infrastructures tend to be found in more economically developed, nominally Christian, countries. There are, however, many notable exceptions to these broad generalizations. For example, Sri Lanka, Kerala (a province of India), Malaysia, Korea, Botswana, Mauritius, Zimbabwe, Barbados, Costa Rica, and Cuba are all noted for their relatively sophisticated and successful public health infrastructures despite their relatively low levels of economic development (Mehrotra, 1997). In addition to being relatively poor, the religious compositions of these countries range across the four predominant faiths. Some of the countries have Christianity as the predominant faith (e.g. Cuba, Zimbabwe), whereas in others Hinduism predominates (e.g. Mauritius); furthermore, for others Buddhism (e.g. Sri Lanka) or Islam (e.g. Malaysia) is in the majority. Third, predominantly Islamic countries have traditional educational systems and beliefs that restrict women’s ability to provide children with medical care (Caldwell, 1986, 1990). In general, Islam has stressed education, but primarily for reading and memorizing the Koran. With some exceptions, contemporary Islamic countries have not stressed education for scientific advancement. In addition, the

238

STEVE CARLTON-FORD

traditional Islamic system stresses education for men rather than for women. Yet it is women’s education that is most strongly related to decreases in child mortality rates (Desai & Alva, 1998). In addition, women’s public contact with men (including physicians) is traditionally restricted. Thus, although there are exceptions, such as Iraq prior to 1991 (Jones, 1999), children in predominantly Muslim countries probably receive poorer health care than in countries of comparable economic development (Caldwell, 1986, 1990). However, traditional Islamic cultural differences do not always disadvantage children’s health. Islamic countries have quite low reported rates of HIV/AIDS (e.g. Al-Sheikh & Fathalla, 1998; Islam et al., 1999). Although the reported rates may be low because of reporting bias (Ridanovic, 1997), the restrictive gender and sexual codes of Islam probably limit the spread of HIV/AIDS through their influence on lifestyle (see Hellmann et al., 1992; Kengeya-Kayondo et al., 1996). As explained above, adult HIV/AIDS places children at risk for higher mortality. Therefore, religion probably is related to children’s mortality through a variety of sometimes causal and sometimes spurious paths, including an association with war and militarization. Thus we must control for differences in the predominant religious culture of societies in order to understand the effects of war on children’s mortality. In summary, war’s influence on children may follow a number of paths which lead to higher mortality. However, potentially offsetting influences (e.g. military participation), as well as exacerbating influences (e.g. religious culture), might lead to a partially spurious relationship of war to child mortality. After discussing our data and measures, we examine the impact of war on children’s mortality while introducing controls for potentially explanatory variables as well as those that may lead to spurious relationships.

DATA AND METHODS The data set was created by merging published country-level statistics from the United Nations International Children’s Emergency Fund (UNICEF), the Stockholm International Peace Research Institute (SIPRI), and from the series World Military and Social Expenditures (Sivard, 1993, 1996). Information is available on the 150 most populous countries in the world. The sample used for analyses consists of eighty-eight developing countries (see Appendix A) for which complete data were available. The basic model (described below) applies equally as well to a larger sample of 137 countries, though results for this larger sample are not reported. Except where explicitly noted, the analyses presented below use the base10 logarithm of each of the measures. Using the log of each measure corrects

The Impact of War, Adult HIV/AIDS, and Militarization

239

for heteroscedacticity and also converts non-linear relationships among variables in the analyses into linear relationships that can be analyzed using OLS regression (McClendon, 1994). In these analyses, regression coefficients represent proportional change; every unit change in an independent variable represents a tenfold increase or decrease in the dependent variable. Data for the measures used in the study are described below. All measures are standardized for population size (see Firebaugh & Gibbs, 1985).

Economic Development We use gross national product per capita (GNP/C) in 1994 expressed in U.S. dollars as our measure of general economic development (from Bellamy, 1996). Information on GNP represents the value of a nation’s “total output of goods and services, valued at current market prices paid by the ultimate consumer” (Sivard, 1993, p. 52). Gross Domestic Product (GDP) or purchasing power parities (PPP) may more accurately reflect the internal economies of countries (Korzeniewicz & Moran, 1997). However, GNP/C correlates very strongly with GDP/C (r > 0.99) and PPP/C (r > 0.95) in our sample and is available for many more countries than these other measures. Preliminary analyses indicated that using more fine-grained measures of the allocation of GNP/C (i.e. to military, education, and health) did not increase explanatory power.

Population Size We use population-standardized (per capita) measures for all of our independent and dependent variables. In conducting such an analysis, one should include the reciprocal of population size (i.e. 1/population) in order to appropriately estimate the intercept (see Firebaugh & Gibbs, 1985). We use the logged reciprocal of total population in 1994 (expressed in millions) as our measure (see Table 1).

Previous Years at War 1946–1989 (War ’46–’89) We use information about whether a country has been involved in either internal or external conflicts that involve 1,000 or more deaths in a year (Sivard, 1996). Countries are coded as involved in armed conflict when they are the primary location of warfare. We then counted the number of years the country was at war from 1946 to 1989 inclusive. Many countries were not at war during this period;

240

STEVE CARLTON-FORD

Table 1. Years at War 1990–1995 Related to Under-5 Mortality, Population, GNP/C, Years at War 1946–1989, Military Participation, and Religious Composition. Years at War 1990–1995 Means for Countries at War Variables

Correlations Between Logged, Standardized Variables

0–2 Years

3–6 Years

Under-5 mortality Logged

85.47 1.76

130.72** 2.03**

0.23*

Total population (millions) Inverse logged

37.74 −0.99

71.95 −1.44***

0.50*

GNP/C Logged

2426.00 3.03

Years at war 1946–1989 Logged, standardized

2.88 0.08

743.00* 2.65**

−0.11

11.51*** 0.18**

0.33**

% Christian Logged

40.52 1.21

31.88 1.05

0.06

% Muslim Logged

33.19 0.97

38.29 1.07

−0.06

% Buddhist Logged

3.83 0.14

8.72 0.20

−0.07

% Hindu Logged

3.24 0.12

3.83 0.15

0.05

19.22 1.00

17.28 0.95

5.66 0.51

3.98 0.43

2.58 −0.40

1.74 −0.49

% Safe water 1990–1995 Logged

66.81 4.14

56.24*

% Women literate 1995 Logged

63.49 1.75

48.72* 1.63*

−0.27**

N

59

29

88

% Other religion Logged Troops per capita 1994 Logged % HIV 1994 Logged

∗ p < 0.05.

∗∗ p < 0.01.

∗∗∗ p < 0.001

(all one tailed tests).

3.93*

0.27** −0.05 0.24** −0.19

The Impact of War, Adult HIV/AIDS, and Militarization

241

others (e.g. Angola, Cambodia, Guatemala) were at war for more than 20 years during this period. Although not completely reliable, Sivard’s estimates are widely cited. To the extent that the variable is unreliable, estimates of its effect should be biased downward, thus providing a conservative analysis of the effects of long-term war.

Current Years at War 1990–1995 (War ’90–’95) We rely on two sources for coding a country as involved in wars during the period 1990 to 1995. First, we use Sivard’s (1996) classification (described just above) of whether a country is at war for the years 1990–1995 inclusive. Second, we use information about major armed conflicts – MACs (SIPRI, 1990, 1991, 1992, 1993, 1994, 1995, 1996). MACs involve a “prolonged combat between the military forces of two or more governments, or of one government and at least one organized armed group, and incurring the battle-related deaths of at least 1,000 people during the entire conflict” (SIPRI, 1996, p. 15). The two different measures overlap in 86% of the cases in our full sample (98 countries not at war; 31 countries at war). Where the two approaches diverge, Sivard’s approach captures conflicts that do not involve well-defined political entities (6%, 9 cases, e.g. Haiti); SIPRI’s approach captures less severe but chronic conflicts (8%, 12 cases, e.g. Israel, Great Britain). Since these chronic armed conflicts outnumber the more severe but less well-defined conflicts (12 to 9), our analysis of the effects of current war on children’s mortality will be conservative. We count the number of years, from 1990 through 1995, that a country was at war. For ease of presentation, we use both a dummy coded and a continuous version of years at war 1990–1995. The dummy coded version contrasts countries that are at war 0–2 years (59 countries, 67% of the analyzed sample) during the period to countries that are at war 3–6 years (29 countries, 33% of the analyzed sample) during that period. We use the continuous logged, population-standardized variable for most of the bivariate and all multivariate analyses.

Children’s Mortality Rates The under-5 mortality rate (U5M) provides an estimate of the likelihood that children will die before reaching age five. Since only young children are included in this measure, the deaths of child soldiers do not confound results. The measure represents the end product of development inputs and is less likely to be affected by the presence of a wealthy minority than many other measures (Bellamy,

242

STEVE CARLTON-FORD

1996, p. 95). The U5M has declined overall in recent decades; yet, considerable between-country differences still remain. The measure ranges from 320 deaths per thousand children in the Niger to below 10 deaths per thousand children in industrialized countries (Bellamy, 1996, 1997). Preliminary analyses using the year-specific mortality rate (the number of under-5 deaths per thousand children under the age of five) provided results similar to those reported below. We use a logged version of the widely publicized UNICEF measure, making our results comparable with previous research (e.g. Bradshaw et al., 1993; Caldwell, 1986, 1990; Pritchett & Summers, 1996).

Military Participation We employ a standard measure of military participation used in previous research, the number of troops per capita. These figures include “regular active-duty forces, including conscripts. Paramilitary forces and reservists are not included” (Sivard, 1996, p. 54, emphasis removed). Presumably these figures include child soldiers. Data on troops per country for 1994 were compiled by Sivard from a wide variety of sources including the UN, NATO, SIPRI, and the CIA (Sivard, 1996, p. 43). We standardized this figure by dividing by overall population size in 1994 to obtain a measure of troops per capita. A logged version of the variable is used in all multivariate analyses.

Religious Composition We obtained information about the religious composition of the countries in our sample from three sources (Crystal, 1997; Gale Research, 1996; Heritage & Cavanagh, 1996). Four figures (see Table 1) are generally available: percent Buddhist, percent Christian, percent Hindu, and percent Muslim. More detailed information about the percent of the population within these categories is not consistently available. Because of the relatively high correlation between percent Christian and percent Muslim, as well as some indication of multi-collinearity in early models, we created a fifth variable-percent other religion. Although we report bivariate results for all of the religious composition variables, we use only the logged versions of percent Buddhist, percent Hindu, percent Muslim, and percent Other Religions in our multivariate analyses. Exclusion of either percent Christian or percent Muslim eliminates problems with multi-collinearity. The results are the same regardless of which variable is excluded.

The Impact of War, Adult HIV/AIDS, and Militarization

243

Women’s Literacy This variable indicates the percent of the adult female population that was literate in 1995. Other research, focusing on long term economic development, has used school enrollment from previous decades as an indicator of education (e.g. Pritchett & Summers, 1996; Shen & Williamson, 1997). We use current female literacy because information is available for more cases (a maximum of 124). Adult female literacy is assumed to be the proximate result of previous schooling. Data were obtained from UNICEF publications (Bellamy, 1997). We use a logged version of the variable in our analyses.

Access to Safe Water For each country with data available, this variable lists the percent of the population that has access to safe water supplies. UNICEF (Bellamy, 1996) provides data collected between 1990 and 1995. We use a logged version of the variable in our analyses. A maximum of 99 cases are available.

Adult HIV Infection This variable represents the World Health Organization’s (WHO) estimate of the percent of the adult population that was infected with HIV as of 1994. WHO uses official country estimates provided by experts or national AIDS programs. Otherwise, they reviewed “HIV sero-prevalence studies, reported AIDS cases, estimates of under-reporting,” as well as other information in making estimates (WHO, 1995). The logged version of the variable is used in all multivariate analyses.

RESULTS Table 1 shows bivariate results for years at war from 1990 through 1995 with the other variables used in the analyses. These results are presented in two forms. Most simply we show a contrast between countries that were at war 0–2 years during the period (the first column) and countries that were at war 3–6 years (the second column). For each variable, we show bivariate results both for variables in their original scales and for the logged, population-standardized variables that are used

244

STEVE CARLTON-FORD

in all subsequent analyses. Finally, the third column shows the bivariate correlations between the logged, population-standardized version of years of armed conflict and the logged, population-standardized version of each other variable used in the multivariate analyses. The results indicate that the under-5 mortality rate is higher the longer a country is at war (85.47 for 0–2 years of war to 130.72 for 3–6 years of war). However, countries at war also differ in a number of ways that might influence the under-5 mortality rate. Overall, countries at war have a lower average level of economic development. Countries at war for three to six years have a lower GNP per capita (US$743) than countries involved in shorter periods of war (US$2,426). These results support previous research which finds that war depresses the economies of countries having conflicts. In addition, countries with more years of war during the early 1990s had already experienced more years of war between 1946 and 1989 (11.51 years compared to 2.88 years). Whether current levels of under-5 mortality are the result of current conflict or the legacies of more long-term conflict is, at this point, open to question. Although countries at war differ economically, the religious composition of these countries in this sample does not vary much with involvement in war. The differences are in the expected direction but are generally not significant in this sample (though they are in the larger sample which is not reported). Countries with few years of war on their territory during the early 1990s have a higher percentage of Christians (40.52) than do countries with more years of war (31.88). In contrast, countries that were at war for longer periods of time have a higher percentage of Muslims (38.29) than do countries with fewer years at war (33.19). Neither the Buddhist nor Hindu composition is significantly related to years of current involvement in war (in either sample). Countries with other religions are somewhat more likely to be involved in armed conflict (r = 0.27). Our bivariate results provide little evidence that predominantly Muslim countries tend to be more involved in armed conflict. Countries with more years of war on their territories have armies that incorporate somewhat, though not significantly, smaller proportions of their populations (r = −0.05). In contrast and consistent with expectations, countries with more years at war have significantly higher rates of HIV infection (r = 0.24), and lower levels of safe water (r = −0.19) and significantly lower levels of women’s literacy (r = −0.27). Arguably, the relationship between Gross National Product per capita and the under-5 mortality rate is so strong that it probably masks the relationship of under5 mortality with other variables. As a result, we present the partial correlations among our variables after controlling for the impact of GNP/C and population size. The matrix of partial correlations in Table 2 indicates that, after these controls, one

The Impact of War, Adult HIV/AIDS, and Militarization

245

Table 2. Partial Correlations for Military Participation, Years at War, and Under-5 Mortality by Religious Composition, HIV, Safe Water, & Women’s Literacy controlling for GNP/C and Population.a

% Buddhist % Christian % Hindu % Muslim % Other religion % HIV 1994 % Safe water 1990–1995 % Women literate 1995

Troops per Capita 1994

Years at War 1990–1995

Years at War 1946–1989

Under-5 Mortality 1995

0.31** −0.35** −0.14+ 0.16+ −0.39*** −0.44*** −0.01 0.07

0.07 −0.09 0.10 −0.04 0.20+ 0.01 −0.09 −0.21+

0.05 0.14+ −0.14 −0.18+ −0.03 −0.11 −0.05 0.12

−0.32** −0.15 −0.19+ 0.27** 0.11 0.24* −0.30** −0.50***

a All variables are logged and standardized for population size. All correlations are based on 88 countries. + p < 0.10. ∗ p < 0.05.

∗∗ p < 0.01.

∗∗∗ p < 0.001.

measure of religious composition – percent other religions – is marginally related to years at war. In addition, religious composition is related to military participation. Countries with higher proportions of Buddhists and Muslims maintain higher levels of troops per capita (r = 0.31 and 0.16 respectively); in contrast, countries with higher proportions of Christians (r = −0.35), Hindus (r = −0.14), and other religions r = −0.39) tend to have significantly fewer troops per capita. After Gross National Product per capita is controlled, religious composition is also related to children’s mortality rates. Countries with higher percentage of either Buddhists or Hindus experience significantly lower under-five mortality (r = −0.32 for Buddhists, and −0.19 for Hindus). The percentage of Muslims is associated with a higher rate of under-five mortality (r = 0.27). Even after controlling for gross national product per capita, higher levels of troops per capita are associated with significantly lower HIV rates (r = −0.44). The adult HIV infection rate is significantly related to higher child mortality rates (r = 0.24). Although infrastructures and women’s education are not strongly related to troops per capita, Table 2 shows their crucial role in affecting children’s mortality: countries whose people have access to safe water and whose women are literate have much lower under-five mortality rates (r = −0.30 for water, and r = −0.50 for literacy). Multivariate analyses are presented in Table 3, which contains four models. First, we present a basic model that estimates the effect of years at war on under-5

246

STEVE CARLTON-FORD

Table 3. Under-5 Mortality as a Function of Religious Composition, HIV, Military Participation, War, Economic Development, and Public Health Infrastructures.a

% Buddhist % Hindu % Muslim % Other religion Troops per capita 1994 Years at war 1990–1995 Years at war 1946–1989 GNP/C 1994 % Safe water 1990–1995 % Women literate 1995 % HIV 1994 % HIV × Troops per capita R2

Model 1 Basic

Model 2 Basic + HIV

Model 3 Basic + Safe Water & Literacy

Model 4 Basic + HIV + Safe Water & Literacy

−0.083 −0.136** 0.079** −0.018 −0.177** 0.855** −0.030 −0.490*** – – – –

−0.100* −0.152** 0.110*** −0.052 −0.045 0.864** −0.045 −0.466*** – – 0.024 0.115*

−0.103* −0.099* 0.014 −0.034 −0.134* 0.503* −0.014 −0.305*** −0.196*** −0.449*** – –

−0.125** −0.098* 0.040 −0.080* −0.002 0.502* −0.018 −0.249*** −0.220*** −0.485*** 0.071* 0.073

0.756

0.789

0.816

0.857

a All

regression coefficients are estimated controlling for the log of the inverse of population. All variables are logged and standardized for population size. All analyses are based on 88 countries. ∗ p < 0.05. ∗∗ p < 0.01. ∗∗∗ p < 0.001.

mortality rates, while controlling for religious composition, military participation, and economic well-being. Model 2 introduces HIV infection as a potential predictor and estimates its interaction with military participation. Model 3 removes HIV (and its interaction with military participation) and introduces access to safe water and women’s literacy as potential predictors. The fourth model considers all of the variables simultaneously. By comparing the effects of war on children’s mortality across these models we gain insight into the processes through which war affects children. In Model 1 (our baseline model) religious composition exerts a small but significant impact on under-5 mortality rates. Countries with higher percentages of Hindus (b = −0.136) have significantly lower under-five mortality rates. In addition, countries with higher percentages of Muslims have somewhat higher under-5 mortality rates (b = 0.079). As expected both troops per capita and GNP per capita have a significant effect in the basic model, reducing the under-five mortality rates (−0.177 for troops in 1994, and −0.490 for GNP in 1994). Years

The Impact of War, Adult HIV/AIDS, and Militarization

247

at War between 1990 and 1995, in contrast, significantly increase the under-five mortality rates (b = 0.855). The number of years at war prior to 1990 has no impact on the 1995 under-5 mortality rate. Model 2 introduces the adult HIV rate and its interaction with troops per capita. Controlling for these two variables leaves the effect of years at war on children’s mortality rates relatively unchanged (b = 0.864). Similarly, controlling for HIV and troop maintenance change the beneficial effect of GNP/C only slightly (b = −0.466). The apparent protective effect of troops per capita is reduced to nonsignificance by controlling for HIV rates and their interaction with troops. The interaction results indicate that a combination of high HIV rates and high troop maintenance is detrimental for children’s mortality rates (b = 0.115). Otherwise, troop maintenance (b = −0.045) and adult HIV rates (b = 0.024) have virtually no impact on children’s mortality rates. In contrast, controlling for the HIV rate and its interaction with troop maintenance heightens the impact of religious composition variables on under-5 mortality. The protective effects of percent Buddhist (b = −0.100), percent Hindu (b = −0.152) and the exacerbating effect of percent Muslim (b = 0.110) are all stronger. Model 3 removes the adult HIV rate (and its associated interaction with troop maintenance) and introduces controls for access to safe water and for women’s literacy. Consistent with previous research, these two variables each independently predict significantly lower levels of under-5 mortality, though women’s literacy exerts a stronger impact (b = −0.449) than does access to safe water (b = −0.196). Controlling for these two variables reduces the impact of both GNP/C (b = −0.305) and years at war (b = 0.503) by about 40%, though each remains significant and in the predicted direction. Controlling for safe water and women’s literacy reduces the effect of troops per capita slightly, but does not eliminate it. The significant protective effects of proportionally large Buddhist (b = −0.103) or Hindu (b = −0.099) populations remain, though these effects are weaker. The effect of percent Muslim becomes non-significant as expected. Model 4 controls for all variables included in the previous three models. Despite these controls, the effect of years of war from 1990 to 1995 on under-5 mortality remains strong (b = 0.502) and significant (at approximately 60% of its baseline value of 0.855). In addition to a significant effect of years at war, a reduced effect of GNP/C (b = −0.249, roughly half of its baseline value), as well as the protective effects of Buddhist (b = −0.125) and Hindu (b = −0.098) remain. A protective effect of other religions (b = −0.080) emerges as significant. Access to safe water (b = −0.220) and women’s literacy (b = −0.485) each significantly reduces under-5 mortality rates, while adult HIV rates increase under-5 mortality rates (b = 0.071). These results remain virtually identical if the non-significant interaction term is removed from the analysis.

248

STEVE CARLTON-FORD

DISCUSSION AND CONCLUSION During the early 1990s between 25 and 35 countries were involved in major armed conflicts each year (Wallensteen & Sollenberg, 1998). The majority of such conflicts were conducted with little international notice. When the conflicts are scrutinized, they received attention primarily because some national interest is involved (e.g. Bosnia and Serbia) or because human rights abuses are particularly egregious (e.g. Rwanda). Little attention is directed to the ways in which armed conflict affects these country’s populations more broadly. We estimated our results in a sample of eighty-eight developing countries, using cross-sectional models. We have, however, reasonable confidence in our findings. The baseline model estimated in a larger sample (n = 137) produced the same pattern of results. We did not examine changes over time for this set of countries because appropriate data are not available for all of the measures. However, the results are consistent with analyses of change over time in a smaller sample (Carlton-Ford, 1999). Those analyses indicate that countries with increasing involvement in armed conflict experience child mortality rates that increase significantly (an average of 60%); countries that decrease their involvement in war show correspondingly strong decreases in their under-5 mortality rates. The research presented here provides strong, consistent evidence that the impact of war on civilian populations extends well beyond those directly affected by the conflict. Specifically, young children die at significantly higher rates in countries involved in major armed conflicts. Detailed analyses of both residuals and predicted values (not shown) indicate that, for the years 1990 through 1995, some countries (e.g. Liberia) suffered up to a ten-fold increase in their under-5 mortality rate as a direct result of armed conflict. Doubling (e.g. Guatemala or Angola) or tripling (e.g. Burundi or Chad) of the under-5 mortality rate is not uncommon. These increases represent a conservative estimate, one that does not take into account the indirect effects on children’s mortality through reduced economic activity because of war or through war’s impact on the education or health infrastructures. If indirect effects were included, the estimates of increased mortality would be much higher. About 40% of war’s total effect was mediated by differences in basic infrastructures, such as women’s education and safe water. Countries with wars on their territory witness the destruction of schools, the disruption of schooling, and the targeting of teachers. All of this decreases the level of educational attainment. We focused on educational attainment of women, the typical care givers for children around the world. Even after controlling for the level of general economic development, lower levels of women’s education were typical of countries at war. Similarly, though less strongly, war is associated with lower levels of access to safe water. Since having both educated women and safe water promote child health, their reduction increases child mortality.

The Impact of War, Adult HIV/AIDS, and Militarization

249

In addition to the effects of war, our baseline analyses replicated previous research showing that countries that maintain large armies relative to the size of their population had lower levels of child mortality. But as expected, this advantage disappeared when large armies were maintained in countries with high levels of HIV infection in the general population. At low levels of HIV or with relatively small militaries there was no effect of either of these variables on children’s mortality. The putative beneficial effect of maintaining troops is completely explained in our models. We find, in contrast to previous speculation, that the putative beneficial effect of military participation is not the result of unanticipated development gains (i.e. in women’s literacy and access to safe water). Rather, we find that a combination of high troop maintenance and high adult HIV infection is associated with higher child mortality, suggesting that previous findings of a positive effect of troop maintenance on child survival were the result of incompletely specified analyses. Also, we found that countries with high proportions of either Buddhists or Hindus in their populations had lower children’s mortality rates than would be expected given their economic and social conditions. Consistent with previous research (e.g. Caldwell, 1990), we found that children in predominantly Muslim countries had higher mortality rates than children in non-Muslim countries with similar general economic conditions. This disadvantage appears to be the result of differences in these countries’ infrastructures. Once we controlled for women’s literacy and access to safe water, the effect of a country’s Muslim composition on children’s mortality rates disappeared. These results support the primary role of public health infrastructure and women’s literacy with regard to children’s mortality. Much of the adverse effect of years at war and half of the beneficial effect of GNP/C is mediated by these variables. Analysts often argue that the complex situations, within which armed conflict is embedded, can only be analyzed case by case. While such an approach does provide a detailed level of understanding about the specific problems that are intertwined in a specific country, it also inadvertently risks minimizing the impact of war. In individual situations, war’s effects may be erroneously attributed by unscrupulous leaders to other aspects of the situation. Without explicit comparison (in this case statistical), war’s effects, as an identifiable aspect of complex situations, may easily be second guessed. Our results provide the first strong statistical evidence that war clearly and demonstrably increases the risk of dying among those individuals in the civilian population – the young children – who can in no way be directly involved in creating or maintaining these ongoing conflicts. Many, although not all, of these conflicts involve national governments; all are potentially amenable to intervention (either before or after conflict starts) by the United Nations. It is time for national and international organizations (both governmental and non-governmental) to consider

250

STEVE CARLTON-FORD

and implement policies that will minimize, if not eliminate, the impact of armed conflict on civilians, especially children. Quantitative results, such as ours, provide information that can be used to change national policy positions either by direct use or through media pressure (see Zwi, 1996). The better we understand how detrimental war is, the process through which armed conflict wreaks its havoc, and the ways in which we may intervene to mitigate those effects, the more likely it is that policy decisions will actually take account of these facts. This paper stands as one small step in the process, similar to that taking place with landmines, of making war obsolete. It is time for sociologists, psychologists, economists, political scientists, and others to use their considerable talents toward that end.

ACKNOWLEDGMENTS Portions of this paper were presented at the annual meetings of the American Sociological Association, August 6, 1999, and at the annual meeting of the Association for Humanist Sociology, November 2000. Research for this paper was supported by a summer grant from the Charles Phelps Taft Foundation and a small research support grant from the Health, Human Values, and Ethics Commission of the Episcopal Diocese of Southern Ohio. The analyses reported and ideas expressed in this paper are the author’s own and do not necessarily reflect the positions of the Taft Foundation or the Episcopal Diocese of Southern Ohio. Many thanks to Robin Carstens, Vallarie Henderson, Philippe Jayat, Jaime McCauley, and Jason Minser, all of whom helped with various aspects of this project. Lisa Cubbins, Robert Garot, and Neal Ritchey all provided encouragement and helpful advice at crucial points. Two anonymous reviewers made very helpful comments on a previous draft of this chapter.

REFERENCES Abu-Amr, Z. (1997). Critical issues in Arab Islamic fundamentalism. In: M. E. Marty & R. S. Appleby (Eds), Religion, Ethnicity, and Self-Identity: Nation in Turmoil (pp. 35–52). Hanover & London: University Press of New England. Adeola, F. O. (1996). Military expenditures, health, and education: Bedfellows or antagonists in third world development? Armed Forces & Society, 22, 441–467. Al-Sheikh, I., & Fathalla, S. E. (1998). Incidence of HIV infection among apparently healthy blood donors in Saudi Arabia (1985–1997). Abstract number 42120 in the International Conference on AIDS, 12, 797.

The Impact of War, Adult HIV/AIDS, and Militarization

251

Anderson, B. A., & Silver, B. D. (1989). Patterns of cohort mortality in the soviet population. Population and Development Review, 15, 471–501. Bellamy, C. (1996). The State of the world’s children. New York: Oxford University Press. Bellamy, C. (1997). The state of the world’s children. New York: Oxford University Press. Bellamy, C. (2000). The state of the world’s children. New York: Oxford University Press. Bhinda, N. (1994). The Kashmir conflict. In: M. Cranna (Ed.), The True Cost of Conflict (pp. 55–80). New York: New Press. Bowen, J. R. (1996). The myth of global ethnic conflict. Journal of Democracy, 7, 3–14. Bradshaw, Y. W., Noonan, R., Gash, L., & Sershen, C. B. (1993). Borrowing against the future: Children and third world indebtedness. Social Forces, 71, 629–656. Bullock, B., & Firebaugh, G. (1990). Guns and butter? The effects of military on economics and social development in the Third World. Journal of Political and Military Sociology, 18, 231–266. Burke, A., & MacDonald, G. (1994). The former Yugoslavia conflict. In: M. Cranna (Ed.), The True Cost of Conflict (pp. 155–196). New York: New Press. Caldwell, J. C. (1986). Routes to low mortality in poor countries. Population and Development Review, 12, 171–220. Caldwell, J. C. (1990). Cultural and social factors influencing mortality levels in developing countries. Annals of the American Academy of Political and Social Science, 510, 44–59. Caldwell, P. (1996). Child survival: Physical vulnerability and resilience in adversity in the European past and the contemporary Third World. Social Science and Medicine, 43, 609–619. Carlton-Ford, S. (1999). The impact of war, international debt, and economic development on children’s mortality rates in developing and transition countries – 1990 to 1995. Presented at the Fourth International Conference on the Child, sponsored by the Organization for the Protection of Children’s Rights, October 13, Laval, Quebec, Canada. Carlton-Ford, S., Hamill, A., & Houston, P. (2000). War and children’s mortality. Childhood, 7, 401–419. Clemens, W. C., & Singer, J. D. (2000). The human cost of war. Scientific American, 282, 56–57. Cockerham, W., & Cohen, L. (1980). Obedience to orders: Issues of morality and legality in combat among U.S. army paratroopers. Social Forces, 58, 1272–1288. Collier, P. (1999). On the economic consequences of civil war. Oxford Economic Papers, 51, 168–183. Collier, P., & Hoeffler, A. (1998). On the economic causes of civil war. Oxford Economic Papers, 50, 563–573. Crystal, D. (1997). The Cambridge Factfinder (1st and 2nd eds). Cambridge, UK: Cambridge University Press. Deger, S. (1985). Human resources, government education expenditure, and the military burden in less developed countries. Journal of Developing Areas, 20, 37–48. Desai, S., & Alva, S. (1998). Maternal education and child health: Is there a strong causal relationship? Demography, 35, 71–81. De Zalduondo, B. O., Msamanga, G. I., & Chen, L. C. (1989). AIDS in Africa: Diversity in the global pandemic. Daedalus, 118, 165–204. Durkheim, E. (1951). Suicide. New York: Free Press. Eckhardt, W. (1990). Civilizations, empires, and wars. Journal of Peace Research, 27, 9–24. Eckhardt, W. (1991). War-related deaths since 3000 B.C. Bulletin of Peace Proposals, 22, 437–443. Firebaugh, G., & Gibbs, J. P. (1985). User’s guide to ratio variables. American Sociological Review, 90, 713–722. Foreman, M. (2002). Combat AIDS: HIV and the world’s armed forces. London: Healthlink Worldwide.

252

STEVE CARLTON-FORD

Gale Research (1996). Countries of the world and their leaders yearbook 1996. New York: Gale Research. Goldstone, J. A., Gurr, T. R., & Moshiri, F. (1991). Revolutions of the late 20th century. Boulder, CO: Westview Press. Gould, P. (1993). The slow plague: A geography of the AIDS pandemic. Oxford, UK: Blackwell. Gurr, T. R., & Moore, W. H. (1997). Ethnopolitical rebellion: A cross-sectional analysis of the 1980s with risk assessments for the 1990s. American Journal of Political Science, 41, 1079–1103. Harris, G. (1997). Estimates of the economic cost of armed conflict: The Iran-Iraq war and the Sri Lankan civil war. In: J. Brauer & W. Gissy (Eds), Economics of Conflict and Peace (pp. 269–287). Aldershot, UK: Avebury. Hellmann, N. S., Grant, R. M., Nsubuga, P. S., Walker, C. K., Kamya, M., Tager, I. B., Jacobs, B., & Mbidde, E. K. (1992). Modifiers of the protective effect of circumcision. International Conference on AIDS, 8(abstract PoC 4299). Heritage, A., & Cavanagh, L. (1996). The world reference atlas. New York: Dorling Publishing. Inglehart, R. (1997). Modernization and postmodernization: Cultural, economic, and political change in 43 societies. Princeton, NJ: Princeton University Press. Islam, M. M., Karim, E., Mian, A. A., Kristensen, S., Chowdhury, M. R., & Vermund, S. H. (1999). An update on the prevalence of HIV/AIDS in Bangladesh. Southeast Asian Journal of Tropical Medicine and Public Health, 30, 246–250. Jones, G. (1999). Iraq-under-five mortality. Retrieved on November 17, 1999 from UNICEF at www.unicef.org/reseval/pdfs/irqu5est.pdf. Kengeya-Kayondo, J. F., Kamali, A., Nunn, A. J., Ruberantwari, A., Wagner, H. U., & Mulder, D. W. (1996). Incidence of HIV-1 infection in adults and socio-demographic characteristics of seroconverters in a rural population in Uganda: 1990–1994. International Journal of Epidemiology, 25, 1077–1082. Kick, E. L., Nasser, R., Davis, B. L., & Bean, L. (1990). Militarization and child mortality in the Third World. Journal of Political and Military Sociology, 18, 285–305. Korzeniewicz, R. P., & Moran, T. P. (1997). World-economic trends in the distribution of income, 1965–1992. American Journal of Sociology, 102, 100–139. Kowalewski, D. (1991). Core intervention and periphery revolution, 1821–1985. American Journal of Sociology, 97, 70–95. Levy, B. S., & Sidel, V. W. (1997). The impact of military activities on civilian populations. In: B. S. Levy & V. W. Sidel (Eds), War and the Public Health (pp. 149–167). New York: Oxford University Press. Low, N., Egger, M., Gorte, A., Sandiford, P., Gonzalez, A., Pauw, J., Ferrie, J., & Smith, G. D. (1993). AIDS in Nicaragua: Epidemiological, political, and sociocultural perspectives. International Journal of Health Services, 23, 685–702. McClendon, M. J. (1994). Multiple regression and causal analysis. Itasca, IL: F. E. Peacock Publishers. Mehrotra, S. (1997). Social development in high-achieving countries: Common elements and diversities. In: S. Mehrotra & R. Jolly (Eds), Development with a Human Face: Experiences in Social Achievement and Economic Growth (pp. 21–61). New York: Oxford University Press. Merli, M. G. (2000). Socioeconomic background and war mortality during Vietnam’s wars. Demography, 37, 1–15. Mielke, J. K., & Pitkanen, K. J. (1989). War demography: The impact of the 1808–1809 war on the civilian population of Aland, Finland. European Journal of Population, 5, 373–398. Mills, C. W. (1959). The power elite. New York: Oxford University Press.

The Impact of War, Adult HIV/AIDS, and Militarization

253

Packard, R. M., & Epstein, P. (1992). Medical research on AIDS in Africa: A historical perspective. In: E. Fee & D. M. Fox (Eds), AIDS: The Making of a Chronic Disease (pp. 346–376). Los Angeles, CA: University of California Press. Preble, E. A. (1990). Impact of HIV/AIDS on African children. Social Science and Medicine, 31, 671–680. Pritchett, L., & Summers, L. H. (1996). Wealthier is healthier. The Journal of Human Resources, 31, 841–868. Ridanovic, Z. (1997). AIDS and Islam. Abstract from Medicinski Arhiv, 51, 45–46. Retrieved on June 23, 2000 at http://hivinsite.ucsf.edu/topics/spirituality/2098.38f2.html. Robinson, I. (1994). The East Timor conflict. In: M. Cranna (Ed.), The True Cost of Conflict (pp. 1–24). New York: New Press. Said, M. (1997). Islam and social conflict. In: M. Kaldor & B. Vashee (Eds), Restructuring the Global Military Sector: New Wars (Vol. 1, pp. 78–107). London: Pinter. Scanlan, S. J., & Jenkins, J. C. (2001). Military power and food security: A cross-national analysis of less-developed countries, 1970–1990. International Studies Quarterly, 45, 159–187. Shen, C., & Williamson, J. B. (1997). Child mortality, women’s status, economic dependency, and state strength: A cross-national study of less developed countries. Social Forces, 76, 667–694. SIPRI (Stockholm International Peace Research Institute) (1990, 1991, 1992, 1993, 1994, 1995, 1996). SIPRI Yearbook. Oxford: Oxford University Press. Sivard, R. L. (1993). World military and social expenditures 1993. Washington: World Priorities. Sivard, R. L. (1996). World military and social expenditures 1996. Washington: World Priorities. Skopol, T. (1979). States and social revolutions. New York: Cambridge University Press. Sorokin, P. (1941). The crisis of our age. New York: Dutton. Sorokin, P. (1942). Man and society in calamity. New York: Dutton. Stouffer, S. A., & DeVinney, L. C. (1955). Who were most critical of the army’s promotion opportunities? In: P. F. Lazarsfeld & M. Rosenberg (Eds), The Language of Social Research (pp. 141–147). Glencoe: Free Press. UNAIDS (1996). Children orphaned by AIDS: Front-line responses from eastern and Southern Africa. Retrieved on January 3, 2001 from the UNAIDS website at http://www.unaids.org/ publications/documents/children/#young. UNAIDS (2002). Orphans and other children affected by HIV/AIDS. Retrieved on February 4, 2004 from the UNAIDS website at http://www.unaids.org/html/pub/topics/youngpeople/uniceforphansfs en pdf.htm. UNAIDS (2003). Children orphaned by AIDS in sub-Saharan Africa. Retrieved on February 4, 2004 from the UNAIDS website at http://www.unaids.org/html/pub/Publications/FactSheets03/FS Orphans Africa en pdf.pdf. UNICEF (2002). HIV/AIDS and children affected by armed conflict. Retrieved on February 4, 2004 from the UNICEF website at http://www.unicef.org/publications/ pub hiv armedconflict en.pdf. Wallensteen, P., & Sollenberg, M. (1989–1997). Armed conflict and regional conflict complexes. Journal of Peace Research, 35, 621–634. World Health Organization (1995). Provisional working estimates of adult HIV prevalence as of the end of 1994 by country. Retrieved from on October 19, 1998 from the WHO website gopher://gpasv05.who.ch/00%5Caidscase%5CHIVTEXT.TXT. Zwi, A. (1996). Numbering the dead: Counting the casualties of war. In: H. Bradby (Ed.), Defining Violence: Understanding the Causes and Effects of Violence (pp. 99–124). Aldershot: Avebury.

254

STEVE CARLTON-FORD

APPENDIX: EIGHTY-EIGHT COUNTRIES INCLUDED IN THE STUDY Afghanistan Algeria Angola Argentina

India Indonesia Iran, Islamic Republic Iraq

Bangladesh Benin Bolivia Botswana Brazil Burkina Faso Burundi

Jamaica Jordan

Cambodia Cameroon Central African Republic Chad Chile China Colombia Congo, Democratic Republic Congo, Republic Cote d’Ivoire Cuba

Kenya Korea, Republic Lebanon Lesotho Liberia Libya

Dominican Republic

Madagascar Malawi Malaysia Mali Mauritania Mauritius Mexico Morocco Mozambique Myanmar

Ecuador Egypt, Arab Republic El Salvador Ethiopia

Nepal Nicaragua Niger Nigeria

Ghana Guatemala Guinea

Oman

Haiti Honduras

Pakistan Panama Papua New Guinea Paraguay

The Impact of War, Adult HIV/AIDS, and Militarization

Peru Philippines Rwanda

255

Togo Trinidad and Tobago Tunisia Turkey

Saudi Arabia Senegal Sierra Leone Singapore South Africa Sri Lanka Sudan Syrian Arab Republic

Uganda United Arab Emirates Uruguay

Tanzania Thailand

Zambia Zimbabwe

Venezuela Vietnam Yemen, Republic

CHANGES IN NONMARITAL COHABITATION AND THE FAMILY STRUCTURE EXPERIENCES OF CHILDREN ACROSS 17 COUNTRIES Jeffrey M. Timberlake and Patrick Heuveline INTRODUCTION Only three decades ago, many demographers believed that the nuclear family – married adults and their biological children – was the modal family structure toward which all societies would rapidly converge (e.g. Goode, 1970). Indeed, during the two decades following World War II, marriage and childbearing in most Western nations tended to: (1) occur early in adulthood; (2) follow a predictable sequence; and (3) be tightly coupled. That is, young couples first married, and then quickly began having children. Over the past few decades in many Western countries, however, marriage and fertility have been increasingly delayed to later adulthood and decoupled from one another, such that the sequence and timing of partnership formation and childbearing have changed dramatically. As a result, most Western nations have experienced increasing rates of out-of-wedlock and out-of-partnership fertility and nonmarital cohabitation1 (as well as divorce) (Goldscheider et al., 2001; Haskey, 2001; Hoem & Hoem, 1988; Kiernan, 2001; Martin & Bumpass, 1989; Murphy, 2000; Noack, 2001; Ostner, 2001; Prinz, 1995; Toulemon, 1997). The pace of change has been so swift that in the preface to the second edition of Marriage, Divorce, Remarriage, Cherlin (1992, p. vii) remarks that only ten years Sociological Studies of Children and Youth Special International Volume Sociological Studies of Children and Youth, Volume 10, 257–278 Copyright © 2005 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1537-4661/doi:10.1016/S1537-4661(04)10013-5

257

258

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

after the publication of the first edition a more appropriate title to the book might have been Cohabitation, Marriage, Divorce, More Cohabitation, and Probably Remarriage. It is certain that recent cohorts of children in the United States and many other Western nations are far more likely to be born out of wedlock or experience the separation of their parents than were the children of previous cohorts. These family changes have sparked popular and scholarly concern over the effects of children’s presumed increased exposure to single-parent families (Amato & Booth, 1997; Blankenhorn et al., 1990; Haveman & Wolfe, 1994; Popenoe, 1988, 1993, 1996; Uhlenberg & Eggebeen, 1986). However, what remains unclear is the extent to which increases in nonmarital cohabitation between children’s biological parents has accounted for (or at least absorbed) the decline in the centrality of marriage as the primary childrearing family structure. In many Western nations there have been large increases in the prevalence of cohabitation in recent years (Casper & Cohen, 2000; Goldscheider et al., 2001; Smock, 2000), and less dramatic increases in the proportion of childhood spent in cohabiting family structures (Bumpass & Lu, 2000; Heuveline, Timberlake & Furstenberg, 2003). This apparent disjuncture is likely due to a combination of at least two factors: first, because cohabiting unions tend to be less stable than marriages (Axinn & Thornton, 1992; Bennett et al., 1988; Hall & Zhao, 1995; Heuveline et al., 2003), then even large increases in the incidence of cohabitation with children may not have substantially offset the decline in the amount of time children can expect to spend with married parents. Second, and more important for the purposes of this article, much of the increase in the prevalence of cohabitation may have occurred when children are not in the household. In this chapter we seek to understand the effects of changes in adult nonmarital cohabitation on the resulting family structure experiences of children. Our primary research question can be stated generally as: to what extent have changes in the incidence and duration of adult cohabitation been translated into changes in children’s exposure to cohabiting-parent, married-parent, and other family structures? More specifically, we ask: for women and children, how much have the expected probability of experiencing cohabitation and the expected duration in a cohabiting family structure changed over time? How much of the increase over time in women’s expected duration in cohabitation occurred while children were co-resident? Finally, how much has the increasing exposure of children to parental cohabitation offset declines in the amount of time children are expected to spend with married biological parents? To address these questions, we analyze partnership and fertility data from 17 industrialized nations. We construct single-decrement and increment-decrement period life tables covering, in most cases, the early 1980s to the early 1990s.

Changes in Nonmarital Cohabitation

259

These tables yield rich cross-national descriptions of expected levels of exposure to cohabitation among women and children. Social scientific research on cohabitation has been limited primarily to the United States, Canada, and a few European nations (Smock, 2000). We extend this line of inquiry to a larger sample of countries, with the goal of advancing our understanding of the extent to which partnership and fertility patterns are converging (or not) across Western nations. Our results provide, to the best of our knowledge, previously unavailable evidence regarding trends in adult cohabitation and children’s family structure experiences across a number of Western nations. Although we do not address questions of the effects of cohabitation on child well-being (Manning & Lichter, 1996), the present analysis is a crucial step in tackling such questions, at least at the national level. We believe that the study of these effects has been hindered by imprecise measurement of childhood family structures, for example, lumping married parents and step-parents together by focusing on the marital status of the head of the household, or assuming the same effect on all children in a given family structure regardless of the time spent in such a family. The present analyses therefore represent a necessary step toward improved subsequent analyses of the societal-level effects of family structures on children.

DATA, MEASURES, AND METHODS Data The data for this chapter come from the Family and Fertility Surveys (FFS), an international sample survey program focusing on partnership formation and fertility issues in the member countries of the United Nations Economic Commission for Europe. The list of participants currently includes over 20 European nations, plus Canada, New Zealand, and the U.S. The FFS program coordinated nationally representative surveys carried out by national statistical offices (Macura & Klijzing, 1992). The first countries participating in the program contributed existing family surveys, while countries joining later attempted to fit the model survey instrument into ongoing data collection efforts (e.g., Cycle 5 of the National Survey of Family Growth in the U.S.). As a result, the years of data collection range from 1988 to 1998, the sampling designs differ, and the questionnaires vary in precise content. The FFS surveys are therefore not perfectly standardized; nevertheless, they constitute a unique source of data on family and fertility trends across countries. More information on the surveys is presented in Kveder (2000), Heuveline et al. (2003), Heuveline and Timberlake (2004), and in Table A1.

260

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

We adopt the “female dominant” approach prevalent in demography by using information from female respondents only to estimate adult and child exposure to cohabitation. Because several of the FFS surveys did not include a male sample, we were able to analyze a larger number of countries with this method. In prior research we found that fewer than 5% of childhood years were expected to be spent not with the mother across countries; therefore, the female dominant approach also yields nearly complete information about the family structure experiences of children (Heuveline et al., 2003, p. 59).

Measures For female respondents to the FFS surveys, we measured transitions into and out of cohabitation, with and without children, from retrospective partnership and fertility histories. For each of up to nine cohabiting relationships, respondents were asked the dates of co-residence (beginning and end), if and how the partnership ended, and the date of marriage, if applicable. In addition, for each of up to 13 live births, respondents reported the date of birth, whether the child was currently in the household, and the date of and reason for departure if he or she was not co-residing at the time of the interview. We reconstructed children’s family structure experiences by combining adult females’ partnership and fertility histories. For each natural-born child, we created an early life course record of living arrangements with the mother and her partners. As long as a child was living with his mother, we knew whether she also lived with a partner and, if so, whether the couple was married. Because we are interested in the extent to which cohabitation acts as a substitute for parental marriage, we accounted for children’s exposure to and duration in three primary states of interest, which we label: (1) “with married parents”; (2) “parental cohabitation (i.e. with cohabiting biological parents)”; and (3) “not with both parents,” a residual state comprising a variety of family structures, such as with a single mother, in a stepfamily, and arrangements not involving the mother (see Heuveline et al., 2003, for more analysis).

Methods Single-decrement Life Tables Period life tables are a general class of demographic models that describe the expected experiences of a cohort of individuals as they transition over time from one life state to another.2 In its classic form, a (single-decrement) mortality life

Changes in Nonmarital Cohabitation

261

table forecasts the dying out of a birth cohort. However, period life tables can describe transitions to and from other life states, such as from single to cohabiting status. For female respondents to the FFS, we used single-decrement techniques to estimate the expected probability of transitioning from single status into a cohabiting relationship by age 453 (in most cases). For children, we calculated the observed percentage of births to cohabiting parents during a given period, and for children not born to cohabiting parents, the expected probability of transitioning into a parental cohabitation by age 1. Increment-decrement Life Tables Because we wish to understand the extent to which cohabitation has accounted for (or at least absorbed) the decline in children’s time spent with married parents, it is crucial to look beyond the mere fact of exposure and analyze the expected duration of children’s exposure to cohabiting parents. Although we are primarily interested in children’s expected exposure to married parents and parental cohabitations, we had to account for the fact that individuals can transition into and out of cohabitation from several family states (e.g. from or to a single mother-headed household), and that individuals may experience multiple occurrences of these transitions over time. Increment-decrement life table techniques enable this kind of analysis by allowing destination states to be “non-absorbing”; that is, in the increment-decrement framework, multiple transitions to (increments) and from (decrements) origin and destination states are possible (Palloni, 2001, p. 256).4 For female respondents to the FFS, we constructed two sets of incrementdecrement life tables: one corresponding to all years of early adulthood, and one corresponding to years in which at least one child was co-resident in the household. We derived our estimates of time spent without children in the household from differences between these two sets of life tables. For children we constructed increment-decrement life tables for all birth statuses. These tables provide the number of person-years that cohorts of women or children would be expected to live in different states if they were exposed to the age-specific transition probabilities observed during the period. We transform these person-year estimates into percentages of young adulthood or childhood by dividing the number of person-years by the age interval (e.g. 4 years expected in cohabitation ÷ 16 years of childhood = 25%). We subsequently use the terms “expected duration” and “childhood expectancy,” analogous to life expectancy in standard mortality life tables, to refer to these percentage estimates. Periods of Observation In order to cover a similar reference period and female age range for all countries, we constructed life tables for many of the countries during a period several years

262

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

before the FFS interview. For example, the Canadian FFS survey was carried out in 1995 (see Table A1), but in the interests of comparability we analyze a three-year period corresponding to calendar years 1991, 1992, and 1993. Although the period of observation varies somewhat across countries, 15 of the 17 periods include at least one of the first four years in the 1990s (Tables 1 and 2). For simplicity of exposition, we subsequently refer to the observed period rates with the shorthand expression “at early 1990s rates.” In our cross-period analyses, we compare the single-period life table estimates with those from a period ranging from 6 to 12 years earlier, depending on the upper age limit of the FFS sample. With these two sets of period life tables, we estimate annualized rates of change in the expected probability of experiencing cohabitation (Table 3), and the expected duration in cohabitation (Table 4).5 For females, we use age 36 in order to estimate trends covering at least eight years, although in Belgium and Hungary data limitations forced us to examine a sixyear period of change and to censor the life table at age 32. In preliminary analysis (not shown), we found that single-period expected probabilities of experiencing a first cohabitation change little between ages 36 and 45; thus, we are confident that right-censoring is not a problem for the cross-period analysis. For children, we were able to maintain an upper age limit of 16 in Austria, Canada, Finland, New Zealand, and Poland. Elsewhere, selectivity concerns compelled us to censor the life table at age 14 (Italy, France, Latvia, and Spain), age 12 (Switzerland), age 10 (Czech Republic, Slovenia, Sweden, and the U.S.), or age 8 (Belgium, Germany, and Hungary).6 Methodological Challenges International research always raises questions about data comparability. The information in the FFS surveys (e.g. dates of birth and co-residence) is fairly objective and therefore less sensitive to the different meanings and interpretations that can hamper comparative research on attitudes, for example. However, there are at least three methodological challenges presented by these data. First, retrospective surveys are subject to recall errors, although dates of birth (of self and own children) and marriage are among the most accurately reported items in such surveys, particularly by women (Poulain et al., 1991). Retrospective data on cohabitation are less reliable, particularly brief cohabitations that occur early in adulthood (Casper & Cohen, 2000; Murphy, 2000). The omission of some early partnerships of older FFS respondents would not bias our single-period estimates because the life table uses information from the youngest respondents to estimate early-adulthood transitions; therefore, these partnerships by construction would have occurred shortly before the FFS interview. However, such omissions would tend to bias estimates of cross-period change (Tables 3 and 4) because

Changes in Nonmarital Cohabitation

263

the estimation of retrospective life tables required using information from older respondents about partnerships that occurred a number of years before the survey. On the other hand, by the very fact of their brevity, the omission of short partnerships would not contribute much bias to duration measures, in either the single- or cross-period analysis. Second, because the FFS surveys were designed to measure marital rather than parental status, female respondents were not directly asked about the identity of the biological fathers of their children. We applied the decision rules described in detail in Heuveline et al. (2003) to distinguish between children living with a biological mother and father and children living with a mother and a male who is not biologically related to the child. Briefly, if a child was born while his or her mother was in a cohabiting partnership (married or not), we assumed that the partner fathered the child. For children born outside of a partnership, we used the timing of the birth and the next union formation to distinguish between a parental union and another partnership. If the next partnership was formed within six months of the birth or if the mother married within a year of the birth, we coded this partner or spouse as the child’s father.7 Any rule based on timing will create some false assignments; however, Heuveline et al. (2003, n. 4) report that the numerical impact of these false assignments is likely to be low because the vast majority of children in these FFS surveys were born within a partnership. Finally, for each child, information was provided on only one departure from the maternal household, if one occurred before the interview. Thus, we had to assume that children continuously co-resided with their mother from birth until either the time of the survey or the date reported as the end of co-residence. Because returns to the maternal household were not reported either, we further assumed that children who were not co-resident at the time of the survey remained out of the maternal household until age 16. It makes some intuitive sense that violations of these assumptions would introduce bias into our estimates of expected duration in a maternal household; however, under stable conditions the expected sum of the two biases is zero.8

FINDINGS Single-Period Analysis Probability of Experiencing Cohabitation We first examine the expected probability of experiencing cohabitation for synthetic cohorts of women and children. The countries in Table 1 are sorted by the percentage of women expected to experience at least one cohabitation by

264

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

Table 1. Expected Probability of Experiencing Cohabitation for Females and Children.a Children Country France Finland New Zealand Sweden Switzerland Austria Canada Germany United States Latvia Slovenia Czech Republic Belgium Hungary Spain Italy Poland

Period 89–91 84–86 90–92 87–89 89–91 90–92 91–93 89–91 91–93 91–93 91–93 93–95 88–90 89–91 91–93 90–92 88–90

Females By Age 44b (1) 83.6 76.2 72.9 66.8 59.9 58.8 55.7 53.5 49.7 47.5 43.8 31.2 28.0 24.0 14.9 9.6 4.4

At Birth (2) 23.8 12.5 18.2 47.4 4.3 17.3 17.4 11.6 11.3 10.6 14.6 9.5 4.0 6.8 5.3 4.1 2.5

From Birth to Age 1 (3) 2.1 0.6 1.9 0.6 0.7 0.7 0.0 1.1 3.3 1.8 1.3 0.9 0.5 0.4 0.1 0.6 0.1

Total by Age 1 (4) 25.9 13.1 20.1 48.0 5.0 18.0 17.4 12.7 14.6 12.4 15.9 10.4 4.5 7.1 5.5 4.7 2.6

Female: Child Ratio (5) = (1)/(4) 3.23 5.82 3.63 1.39 11.92 3.27 3.21 4.21 3.41 3.83 2.76 2.99 6.29 3.38 2.72 2.04 1.71

Note: Table sorted in descending order by column 1. a Data from the Family and Fertility Surveys. b See Table A1 for upper age limits.

age 45 (except where noted). At early 1990s rates, a majority of women can expect to form at least one cohabitation in eight countries (France, Finland, New Zealand, Sweden, Switzerland, Austria, Canada, and Germany), ranging from 83.6% in France to 53.5% in Germany. The United States follows with just under 50%. Five other countries – Latvia, Slovenia, Czech Republic, Belgium, and Hungary – show percentages between one-quarter and one-half of a cohort of women. In Spain, Italy, and Poland, cohabitation is a fairly rare occurrence, with fewer than 15% of a cohort of 18 year-old women expected to experience cohabitation.9 In general, the expected probability that children will experience parental cohabitation reflects the probability for women (the correlation between column 1 and 4 is 0.65). Children’s expected probabilities of experiencing parental cohabitation range from nearly half of a birth cohort in Sweden (48.0%) to fewer than 3% in Poland. There appears to be have been a fair amount of convergence over time in children’s expected probability of experiencing parental cohabitation, with 9 of the 17 countries exhibiting probabilities ranging from 10 to 20% of a birth cohort (New Zealand, Austria, Canada, Slovenia, U.S., Finland, Germany, Latvia,

Changes in Nonmarital Cohabitation

265

and Czech Republic). Fewer than 10% of a birth cohort of children in Belgium, Hungary, Italy, Spain, and Poland can expect to experience parental cohabitation during childhood. Column 5 of Table 1 demonstrates that nonmarital cohabitation is much more prevalent as a pre-childbearing (or post-childrearing) family form than as a family structure to which children are exposed. The ratio of the percentage of women expected to experience cohabitation by age 45 to the percentage of children expected to experience a parental cohabitation (column 7) exceeds 2.0 in all countries except Sweden and Poland, and approaches or exceeds 3.0 in 12 of the 17 countries. The most extreme case is Switzerland, where about 60% of women are expected to experience cohabitation, while only 5% of children are expected to experience a parental cohabitation. Thus, women in Switzerland are about 12 times more likely to experience cohabitation than children are to experience a parental cohabitation, at early 1990s rates. These results – high adult prevalence and low child exposure – are consistent with previous research on the Swiss family system (Charton & Wanner, 2001). Expected Duration in Cohabitation The countries in Table 2 are sorted by the total percentage of early adulthood (ages 18–45 in most countries) women were expected to spend in a cohabitation (column 3), again at early 1990s rates. Although Sweden ranked behind France, Finland, and New Zealand in terms of the expected probability of female cohabitation (Table 1), Sweden stands out in Table 2 with nearly a third of the time between age 18 and 42 expected to be spent in cohabitation (about 7.8 years). French women can expect to spend about a quarter of their early adulthoods in cohabitation (6.7 years from age 18 to 45), while women in New Zealand, Austria, and Finland are expected to spend about one-fifth of early adulthood in cohabitation. Eight other countries show expected durations of 9–18%, while in Hungary, Spain, Italy, and Poland, women’s expected duration in cohabitation is less than 7% of early adulthood. Column 5 of Table 2 presents children’s expected durations from birth to age 16 in parental cohabitation. Column 5 shows that only in France are children expected to spend more than 10% of childhood in parental cohabitation. Five more countries exhibit childhood expectancies of 6–9% (about 1.0–1.4 child-years). Column 4 shows that in most countries, children’s overwhelming experience is with married parents, at least at early 1990s rates. Exceptions include Latvia, the U.S., New Zealand, Canada, Austria, and Germany, where children can expect to spend over a quarter of childhood not with both parents. Prior research has found that the majority of this time not with both parents is spent with a single mother, rather than in other two-parent family structures (Heuveline et al., 2003, p. 59).

266

Table 2. Expected Duration in Cohabitation for Females and Children.a Females (% of Age 18–44b ) Period 87–89 89–91 90–92 90–92 84–86 91–93 89–91 91–93 91–93 89–91 93–95 88–90 91–93 89–91 91–93 90–92 88–90

With Children (1) 15.5 12.0 10.3 8.7 5.6 11.0 4.0 7.6 7.7 1.3 5.4 1.8 5.1 2.7 2.0 1.6 1.5

Without Children (2) 17.1 13.7 11.0 11.2 13.1 6.6 11.5 5.6 4.3 10.5 5.5 7.7 4.1 4.1 2.3 2.0 0.3

Total (3) 32.5 25.7 21.3 19.9 18.6 17.5 15.5 13.2 11.9 11.8 10.8 9.5 9.2 6.8 4.3 3.6 1.7

With Married Parents (4) 70.3 69.0 61.3 65.9 83.4 61.6 68.6 83.5 60.3 89.2 75.4 89.3 62.5 84.9 90.3 92.9 84.4

With Cohabiting Parents (5) 8.0 11.7 6.9 6.3 3.7 8.7 4.2 6.1 3.0 0.8 3.0 2.4 2.6 3.1 2.6 2.2 2.8

Note: Table sorted in descending order by column 3. from the Family and Fertility Surveys. b Age 38 in Belgium and Hungary, 40 in Germany, and 42 in the Czech Republic, Sweden and the U.S. c Age 14 in Belgium and Hungary. a Data

Not With Both Parents (6) 21.6 19.3 31.8 27.8 13.0 29.7 27.2 10.4 36.6 10.1 21.6 8.3 34.8 12.0 7.1 4.9 12.8

Female: Child Ratio (7) = (3)/(5) 4.04 2.20 3.07 3.14 5.04 2.02 3.65 2.17 3.92 15.68 3.64 3.94 3.50 2.23 1.64 1.66 0.63

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

Country Sweden France New Zealand Austria Finland Canada Germany Slovenia Latvia Switzerland Czech Republic Belgium United States Hungary Spain Italy Poland

Children (% of Birth to Age 16c )

Changes in Nonmarital Cohabitation

267

Fig. 1. Expected Duration in Cohabitation for Females, by Presence of Children. Note: See Table 2 for period of analysis. * Age 38 in Belgium and Hungary, 40 in Germany, and 42 in Czech Republic, Sweden and the U.S. Source: Data from the Family and Fertility Surveys.

The correlation between columns 3 and 5 is 0.78, again suggesting that children live with cohabiting parents to a similar extent as this partnership form is culturally sanctioned for adults. However, as in Table 1, column 7 in Table 2 shows that women in all countries except Poland are expected to spend a much longer relative duration in cohabitation than are children – in all countries except Spain, Italy, and Poland, the female to child ratio in column 8 approaches or exceeds 2.0. Switzerland again stands out, with a ratio of nearly 16. Furthermore, as demonstrated in columns 1 and 2 of Table 2 and in Fig. 1, in most countries the percentage of early adulthood spent cohabiting without children exceeds the percentage spent with children. These findings suggest that the role of parental cohabitation in replacing parental marriage as a childrearing institution may be overstated in current popular and scholarly thought. They further lead us to hypothesize that much of the increase in adult cohabitation over time has occurred when children are not in the household. In the next section we examine this hypothesis by analyzing patterns of change across countries.

Cross-Period Analysis Change in the Probability of Experiencing Cohabitation We now shift focus from single-period, cross-national comparisons to withincountry trends over time. The countries in Table 3 are sorted by the annualized rate

268

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

Table 3. Annualized Rates of Change in the Expected Probability of Experiencing Cohabitation for Females and Children.a

Country Czech Republic Hungary Belgium Latvia Germany Spain Italy Canada Finland Slovenia United States France Poland New Zealand Austria Sweden Switzerland

Time Intervalb (Years) 8 6 6 8 8 8 8 12 10 8 8 8 10 12 12 8 8

Annualized Rate of Change (in % Per Year) Females By Age 44c (1) 13.1 10.0 8.0 7.3 6.9 6.6 5.6 4.6 4.6 4.4 3.8 3.5 2.6 2.4 2.2 2.1 −0.5

Children by Age 1 (2) 15.0 12.9 11.4 4.9 2.1 12.6 12.6 12.2 3.2 3.7 5.0 11.6 7.2 5.0 7.1 2.5 1.4

Female: Child Ratio (3) = (1)/(2) 0.87 0.78 0.70 1.49 3.26 0.52 0.45 0.38 1.41 1.21 0.76 0.30 0.37 0.48 0.31 0.83 −0.35

Note: Table sorted in descending order by column 1. a Data from the Family and Fertility Surveys. b Refers to the beginning of a three-year interval prior to the beginning of the three-year interval noted in Table 1. c Age 32 in Belgium, Germany, and Hungary, and age 34 in Czech Republic, Sweden, and the U.S.

of change (see footnote 5) from time t – n to t in females’ expected probability of experiencing a cohabitation by age 36 (except where noted). The findings in Table 3 indicate that the expected probability of forming a cohabitation was increasing in most countries in the years prior to the FFS interviews, in some places quite rapidly. Nine countries experienced increases of about 5% per year or greater, including remarkably swift changes in Czech Republic (13.1% per year) and Hungary (10% per year). Rapid increases were seen in many low-prevalence countries, including Hungary, Belgium, Spain, and Italy; however, even Germany, Finland, the U.S., and France experienced fairly robust annualized rates of change. Four of the highest-prevalence countries from Table 1 – New Zealand, Austria, Sweden, and Switzerland – were experiencing much slower or even negative growth, suggesting that by the early 1990s, these countries had reached a plateau in terms of the contribution of nonmarital cohabitation to family formation.

Changes in Nonmarital Cohabitation

269

Column 2 of Table 3 shows uniformly positive annualized rates of change in children’s probability of experiencing a parental cohabitation. As with adult change, high rates of change were observed in countries that had fairly low singleperiod probabilities of childhood exposure, including Czech Republic, Belgium, Hungary, Italy, and Spain. In addition, however, France and Canada experienced some of the highest rates of change, suggesting that the popularity of rearing children within cohabitation has grown rapidly in these countries. Annualized rates of change of between 3 and 7% per year were also observed in Austria, New Zealand, Latvia, Slovenia, and Finland. Growth was lowest in Sweden, Germany, and Switzerland, suggesting more stability in these countries family formation regimes. In Sweden, this indicates stability in a system in which both adult and child exposure to cohabitation is high, while in Switzerland, the previously-noted pattern of high adult and low child exposure appears to remain in place. The ratios in column 3 indicate that children’s exposure to parental cohabitation is increasing faster than overall exposure to cohabitation for women. Only four countries (Latvia, Germany, Finland, and Slovenia) had ratios greater than 1 (plus Switzerland, where adult growth was negative). This finding is largely due to the fact that children’s exposure to parental cohabitation is still much rarer than females’ exposure (Table 1); thus, identical absolute increases (or even smaller absolute increases for children) would yield greater annualized rates of change for children. Change in Expected Duration in Cohabitation In Table 4 we assess the magnitude and direction of changes in women and children’s expected duration in cohabitation. The countries in Table 4 are sorted by change in females’ total expected duration in cohabitation (column 3). As in Table 3, the figures are in annualized rates of change in expected duration in cohabitation from age 18 to 36 for women, except where noted, and from birth to the upper age limit for children. At the time of the interview, increases of about 5% per year or more were occurring in all countries except the United States, Switzerland, Germany and Sweden. In Sweden, which had the highest female incidence of exposure to cohabitation and longest expected duration as of the early 1990s, expected duration in cohabitation for females was declining slightly at the time of the survey. In 7 of the 17 countries, increases without children were greater than increases with, with the reverse being true in the other 10. The total increase in expected duration in cohabitation was dominated by increases without children in Latvia and Slovenia, suggesting that nonmarital cohabitation was becoming more socially acceptable in these countries at the time of the FFS interview, provided that children were not in the household. In France, Spain, Hungary, New Zealand, Canada, and Belgium, increases in female cohabitation with children were

270

Table 4. Annualized Rates of Change in Expected Duration in Cohabitation for Females and Children.a Females (in % of Age 18 18 to Upper Age Limit Per Year

Children (in % of Birth to Upper Age Limit Per Year)

Notes: Table sorted in descending order by column 3. See Table 3 for intervals and child upper age limits. a Data from the Family and Fertility Surveys. b Refers to the beginning of a three-year interval prior to the beginning of the three-year interval noted in Table 1.

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

Time Female: Upper Age Limit Intervalb With Without With Married With Cohabiting Not With Child Ratio Country (Years) Females Children Children (1) Children (2) Total (3) Parents (4) Parents (5) Both (6) (7) = (3)/(5) Czech Republic 8 34 10 13.4 15.0 14.1 −1.7 28.6 7.2 0.49 Poland 10 36 16 14.1 9.9 13.2 −0.5 16.1 2.0 0.82 Latvia 8 36 14 9.5 15.6 11.7 −2.3 5.1 5.0 2.31 Finland 10 36 16 6.9 7.9 7.7 −0.2 0.6 0.9 13.60 Slovenia 8 36 10 5.4 10.4 7.4 −0.1 5.2 −2.9 1.40 France 8 36 14 10.2 3.4 5.9 −0.8 6.6 1.2 0.89 Spain 8 36 14 12.8 2.4 5.7 −0.4 20.7 1.4 0.28 Hungary 6 32 8 9.0 1.9 5.6 −0.4 15.0 −1.4 0.37 New Zealand 12 36 16 9.3 3.0 5.5 −1.6 5.1 3.0 1.08 Canada 12 36 16 9.0 1.9 4.9 −2.7 7.1 8.4 0.69 Austria 12 36 16 5.1 4.4 4.7 −0.9 5.9 1.4 0.80 Belgium 6 32 8 10.5 3.6 4.7 −0.4 12.6 3.6 0.37 Italy 8 36 14 6.2 3.4 4.5 −0.3 6.4 5.0 0.70 United States 8 34 10 3.6 1.4 2.4 0.1 −0.2 −0.3 −13.41 Switzerland 8 36 12 −1.7 2.2 1.8 −0.4 4.8 2.8 0.37 Germany 8 32 8 −0.4 1.3 0.8 −0.4 −2.6 2.2 −0.31 Sweden 8 34 10 −1.6 0.3 −0.5 −0.2 0.2 0.6 −2.23

Changes in Nonmarital Cohabitation

271

occurring much faster than without, suggesting that cohabitation was increasingly being used and accepted as a childrearing family form. Table 4 also presents data on changes in the share of childhood spent in different family structures over time. In all countries except the U.S., childhood expectancy with married parents was declining slowly in the years before the FFS survey, and in all countries except the U.S. and Germany, childhood expectancy with cohabiting parents increased. Increases in childhood expectancy with cohabiting parents were largest in Czech Republic (28.6% per year), Spain (20.7%), Poland (16.1%), and Belgium (12.6%). These were countries with among the lowest childhood expectancies in cohabitation (see Table 2, column 5); thus, small absolute increases over this low baseline resulted in large annualized rates of change. The countries with higher childhood expectances in Table 2 were still experiencing fairly rapid growth at the time of the FFS surveys, including 7.1% per year in Canada, 6.6% per year in France, 5.9% per year in Austria, 5.2% per year in Slovenia, and 5.1% per year in New Zealand. Sweden, which had the highest childhood expectancy at the time of the survey, was experiencing almost no growth in children’s expected duration in parental cohabitation. As in Table 3, rates of change in childhood expectancy with cohabiting parents generally exceeded rates of change in adult expected duration in cohabitation (column 7). However, this is again largely due to the fact that single-period childhood expectancy with cohabiting parents was quite low in most countries (Table 2). It is important to note that although the annualized rates of change for childhood expectancy with cohabiting parents appear impressive compared to rates of decline in time with married parents (columns 5 and 4, respectively), increases in time spent with cohabiting parents were not sufficient to completely absorb the decline in time spent with married parents. Note that in all countries except Slovenia and Hungary, childhood expectancy not with both parents was increasing at the time of the FFS survey. This indicates that even substantial increases in expected duration in parental cohabitation were not sufficient to prevent children from spending an increasing share of childhood without one or both biological parents. In most countries, this has meant increasing exposure to single-parent households (Heuveline et al., 2003).

SUMMARY AND CONCLUSIONS Although the countries we analyzed display substantial variation both in the crosssection and over time, three basic conclusions emerge. First, at early 1990s rates, in most countries the expected probability of women experiencing a cohabitation during early adulthood was substantially larger than the probability of children

272

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

experiencing a parental cohabitation (Table 1, column 5). Likewise, the relative expected duration in cohabitation was longer for women than for children (Table 2, column 7). Moreover, Fig. 1 shows that in most countries, women’s expected duration in cohabitation without children exceeds that with children. These findings indicate that cohabitation for many women continues to be a non-childrearing living arrangement. Second, the probability of women experiencing a cohabitation was increasingly fairly rapidly in most countries (column 1 of Table 3). For children, we found that the probability of experiencing a first cohabitation with biological parents was increasing even faster in most countries (column 3). Third, we found that in most countries, females’ expected duration in cohabitation with children was increasing faster than expected duration without children (columns 1 and 2 of Table 4), although there was substantial cross-national variation in these rates of growth. Latvia and Slovenia appeared to be countries in the early stages of transition to a more prominent role of cohabitation in childrearing. In these countries, the rate of growth in expected duration in cohabitation without children was much larger than that with children. In contrast, France was perhaps the most notable example of a country in which childrearing within cohabitation was growing much faster than childless cohabitation. Change in children’s expected duration with cohabiting parents also displayed great variation, with dramatic annualized rates of positive growth experienced in countries where children have historically spent very little time with cohabiting parents (e.g. Spain, Hungary, and Belgium). Countries with higher single-period childhood expectancies in parental cohabitation (Austria, Canada, France, New Zealand, and Slovenia) also experienced fairly robust annualized rates of change in expected duration with cohabiting parents, with the notable exception of Sweden. Sweden experienced little change in either adulthood or childhood expectancy in cohabitation. This may be because the role of cohabitation in family formation has to some extent reached an equilibrium in Sweden, with those couples who wish to pursue this family formation strategy given sufficient cultural and institutional supports to do so. As Kiernan (2001, p. 3) argues, Sweden may have reached a stage in which cohabitation is “indistinguishable from marriage.” We suspect that cross-national variation in trends in cohabitation is due in large part to differential rates of the relaxation of cultural norms against unmarried adults: (1) living together without children; (2) bearing children together; and (3) raising children together. Our argument proceeds as follows: In the two decades following World War II, the duration between the age of consent and marriage was short and dominated by single (i.e. non-cohabiting) status in Western societies. In addition, the predominant family formation sequence was singlehood, marriage, and childbearing, with marriage and childbearing tightly coupled, that is, first birth

Changes in Nonmarital Cohabitation

273

occurring soon after marriage. Whereas cohabitation until at least 1970 was almost exclusively reserved for the poor and the divorced (Kiernan, 2001, p. 2), in recent decades, many Western societies have increasingly sanctioned the formation of a pre-marital cohabiting partnership for a heterogeneous group of young adults. However, as marriage and especially fertility have been postponed to later adulthood, young couples now face a longer period of time in limbo between the age of partnership formation and first birth. Thus, additional delays in marriage and fertility haven’t dramatically affected the incidence of cohabitation formation for young women (except in low-prevalence countries such as Czech Republic, Hungary, Italy and Spain); that is, the expected probability of ever experiencing a cohabitation hasn’t been much affected by whether the duration from age 18 to marriage or first birth is 5 years or 10 years. However, because the amount of time spent without children is now much longer, expected duration in cohabitation has increased more appreciably for women. In addition, although childrearing within marriage is still the overwhelming norm in Western societies (even in Sweden – compare columns 4 and 5 of Table 2), the precise sequence of marriage and childbearing is becoming more flexible. Thus, children have become much more likely to experience parental cohabitation. Relative to a low baseline, they are experiencing a relatively longer share of childhood in cohabitation; however, in absolute terms the change has not been substantial in most countries (Heuveline et al., 2003, p. 61). We conclude by attempting to provide an answer to the question implied by the title of this chapter: how much have changes in adult nonmarital cohabitation affected the family structure experiences of children? Our answer is: some, but probably not as much as is commonly believed. That is, despite some rapid changes from roughly the mid-1980s to early 1990s, adult females were, as of the early 1990s, still much more likely than children to experience cohabitation. Furthermore, women’s expected duration in cohabitation without children was substantially higher than that with children, and in only one country (France) could children expect to spend more than 10% of childhood with cohabiting parents. Finally, despite rapid increases in children’s expected duration in parental cohabitation, childhood expectancy not with both biological parents was also on the rise. On the other hand, we observed quite rapid rates of change for children in a number of countries; thus, it is conceivable that after another ten years of change (from the early 1990s to the early 2000s), children’s family structure experiences might be considerably influenced by changes in the social acceptability of nonmarital cohabitation. The trends documented here pertain to what demographers, following van de Kaa (1987), commonly refer to as the “second demographic transition.” While nonmarital cohabitation, out-of-wedlock fertility, and divorce appear to be on

274

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

the increase in all countries, the pace of the change has varied across countries, as have the consequences for children’s experience of family structures. The second demographic transition can likely be traced to some common macro-level factors affecting all Western countries, but these results also suggest the presence of idiosyncratic national factors that affect the pace of changes in adult family behavior and the way they translate in changes for children. Comparative studies of these macro-level factors that mediate the effect of family changes on children’s experiences are as important as studies of the individual-level predictors of family behavior. Although a wealth of individual survey data have benefited the latter program of research, comparative databases on national characteristics that may affect family decisions have not been developed as rapidly and should now be prioritized.

NOTES 1. For the purposes of this chapter we define “cohabitation” as “a spell of co-residence of at least one month in duration by two unmarried, romantically involved, heterosexual adults.” For some research on the prevalence and trends in the legal recognition of same-sex partnerships, see Black et al. (2000) and Bradley (2001), respectively. 2. In contrast to cohort life tables, which simply record what happens to a cohort as it ages, period life tables estimate what would happen to a cohort if it were to experience the age-specific transition probabilities that existed during the period in which the cohort is defined (e.g. children born from January 1, 1991 to December 31, 1993, or women turning age 18 during this interval). Because the future experiences of such cohorts are not observed, but rather estimated with period conditions, demographers refer to them as “synthetic” cohorts (Preston et al., 2001, p. 42). 3. “By age 45 (or 36 or 16, etc.)” may be interpreted as “by exact age 45 (or 36, or 16, etc.).” That is, “experiencing cohabitation by age 45” refers to transitions into cohabitation by age 44 and 364 days. 4. See Heuveline and Timberlake (2002) and Heuveline et al. (2003) for details on increment-decrement life table construction using FFS data. 5. These rates of change follow the formula: 
 Xt 1 rt−n,t = 100 ln , n Xt−n where n equals the number of years in the interval, Xt equals the life table estimate for the most recent period, and Xt−n equals the life table estimate from the period n years back in time. 6. By “selectivity concerns,” we are referring to the fact that the children in older life tables were on average born to younger mothers than the children in more recent life tables. Because mothers under age 25 are more at risk for out-of-wedlock childbearing (Morgan & Rindfuss, 1999) and partnership separation, we had to censor certain countries’ life tables to ensure that not all of the children in them were born to mothers younger than age 25.

Changes in Nonmarital Cohabitation

275

7. This procedure could not be carried out in Canada because all “month” variables (e.g. of partnership beginning, marriage, partnership end, children’s births) were coded to the same month within respondent. Thus, the first transition a child could experience after birth (for example, the marriage of his cohabiting (at birth) biological parents) could by construction only occur one full year after birth. As a result, our timing rules could not code this first transition as one into parental cohabitation. This will exert downward bias in our estimates of Canadian children’s exposure to parental cohabitation; however, given that the average expected probability of exposure to parental cohabitation after birth is less than 1% in the other 16 countries we analyze in this article, the magnitude of the bias in Canada is likely to be small. 8. Heuveline et al. (2003, n. 3) provide empirical evidence to support this claim. More formally, suppose a child lives with her mother from birth to age a, then departs the maternal household until age b, at which point she returns, where birth < a < b < 16. Let A = the interval from birth to age a, B = the interval from age a to b, and C = the interval from age b to 16. If the FFS survey is administered during A, then there is no bias in our estimate of time spent in a maternal household. If it is administered during B, then our estimate is downwardly biased by C. If it is administered during C, then our estimate is upwardly biased by B. Thus, under stable conditions, Bias = E(A[0] + B[−C] + C[B]) = 0 + E(B[−C]) + E(C[B]) = 0. 9. In other research (Heuveline & Timberlake, 2004) we use the overall incidence of cohabitation as a key indicator of the “role” cohabitation plays across countries. Briefly, we argue that although cross-national differences in individual characteristics largely explain variation in the prevalence of particular sequences of partnership formation, marriage, childbearing, and childrearing, national-level differences also depend on the cultural norms and institutional supports that families encounter in each society. We argue that these country-specific effects can be detected in measures of central tendency on indicators such as the incidence of, duration of exposure to, and route of exit from cohabitation for adults and children.

ACKNOWLEDGMENTS A previous version of this chapter was presented at the annual meeting of the American Sociological Association, Chicago, IL, August 2002. This research has been supported in part by Grant 2 P30-HD18288-16 from the National Institute of Child Health and Human Development (NICHD), and a pre-doctoral fellowship from the Population Research Center at the University of Chicago, NICHD Grant 2 T32-HD073-02. The authors wish to thank the Advisory Group of the Fertility and Family Surveys (FFS) programme of comparative research for its permission, granted under project number 57, for use of FFS data for this study. Thanks also to the Institut National de la Statistique et des Etudes Economiques, the Bundesinstitut f¨ur Bev¨olkerungsforschung, the Swiss/Federal Statistical Office,

276

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

and Statistics Sweden for additional releases and support. Larry Bumpass, Martine Corijn, Gert Hullen, Erik Klijzing, Kurt Schmidheiny, and Laurent Toulemon provided essential information about the survey data, and Ye Luo created the working files used in the analyses presented here. Finally, thanks to Loretta Bass, Sarah Beth Estes, Annulla Linders, Jennifer Malat, and two anonymous reviewers for their comments on an earlier draft.

REFERENCES Amato, P. R., & Booth, A. (1997). A generation at risk: Growing up in an era of family upheaval. Cambridge, MA: Harvard University Press. Axinn, W. G., & Thornton, A. (1992). The relationship between cohabitation and divorce: Selectivity or causal influence? Demography, 29, 357–374. Bennett, N. G., Blanc, A. K., & Bloom, D. E. (1988). Commitment and the modern union: Assessing the link between premarital cohabitation and subsequent marital stability. American Sociological Review, 53, 127–138. Black, D., Gates, G., Sanders, S., & Taylor, L. (2000). Demographics of the gay and lesbian population in the United States: Evidence from available systematic data sources. Demography, 37, 134–159. Blankenhorn, D., Bayme, S. & Elshtain, J. B. (Eds) (1990). Rebuilding the nest: A new commitment to the American family. Milwaukee, WI: Family Service America. Bradley, D. (2001). Regulation of unmarried cohabitation in west-European jurisdictions: Determinants of legal policy. International Journal of Law, Policy and the Family, 15, 22–50. Bumpass, L. L., & Lu, H.-H. (2000). Trends in cohabitation and implications for children’s family contexts in the United States. Population Studies, 54, 29–41. Casper, L. M., & Cohen, P. N. (2000). Does the POSSLQ measure up? Historical estimates of cohabitation. Demography, 37, 237–245. Charton, L., & Wanner, P. (2001). La premi`ere mise en couple en Suisse: Choix du type d’union et devenir de la cohabitation hors mariage [First couple formation in Switzerland: Choice of union type and outcome of extra-marital cohabitation]. Population, 56, 539–568. Cherlin, A. J. (1992). Marriage, divorce, remarriage (2nd ed.). Cambridge, MA and London: Harvard University Press. Goldscheider, F., Turcotte, P., & Kopp, A. (2001). The changing determinants of women’s first union formation in industrialized countries: The United States, Canada Italy and Sweden. Genus, 57, 107–113. Goode, W. J. (1970). World revolution and family patterns. New York: Free Press. Hall, D. R., & Zhao, J. Z. (1995). Cohabitation and divorce in Canada: Testing the selectivity hypothesis. Journal of Marriage and Family, 57, 421–427. Haskey, J. (2001). Demographic aspects of cohabitation in Great Britain. International Journal of Law Policy and the Family, 15, 51–67. Haveman, R., & Wolfe, B. (1994). Succeeding generations: On the effects of investments in children. New York: Russell Sage Foundation. Heuveline, P., & Timberlake, J. M. (2002). Toward a child-centered life course perspective on family structures: Multistate early life tables. In: E. Klijzing & M. Corijn (Eds), Fertility and Partnership in Europe: Insights and Lessons from Comparative Research (Vol. II). New York/Geneva: United Nations.

Changes in Nonmarital Cohabitation

277

Heuveline, P., & Timberlake, J. M. (2004). The role of cohabitation in family formation: The United States in comparative perspective. Journal of Marriage and Family, 66, 1214–1230. Heuveline, P., Timberlake, J. M., & Furstenberg, F. F., Jr. (2003). Shifting childrearing to single mothers: Results from 17 western countries. Population and Development Review, 29, 47–71. Hoem, B., & Hoem, J. M. (1988). The Swedish family: Aspects of contemporary developments. Journal of Family Issues, 9, 397–424. Kiernan, K. E. (2001). The rise of cohabitation and childbearing outside marriage in western Europe. International Journal of Law, Policy and the Family, 15, 1–21. Kveder, A. (2000). Data quality issues in the scope of international comparison in FFS surveys. Paper Presented at the FFS Flagship Conference: Partnership and Fertility – A Revolution? Brussels, Belgium, May 2000. Macura, M., & Klijzing, E. (1992). Fertility and family surveys in countries of the ECE region. Manuscript, Population Activities Unit, Economic Commission for Europe, Geneva. Manning, W., & Lichter, D. (1996). Parental cohabitation and children’s economic well-being. Journal of Marriage and Family, 58, 998–1010. Martin, T. C., & Bumpass, L. L. (1989). Recent trends in marital disruption. Demography, 26, 37–51. Morgan, S. P., & Rindfuss, R. R. (1999). Reexamining the link of early childbearing to marriage and to subsequent fertility. Demography, 36, 59–75. Murphy, M. (2000). The evolution of cohabitation in Britain, 1960–95. Population Studies, 54, 43–56. Noack, T. (2001). Cohabitation in Norway: An accepted and gradually more regulated way of living. International Journal of Law, Policy and the Family, 15, 102–117. Ostner, I. (2001). Cohabitation in Germany: Rules, reality and public discourses. International Journal of Law, Policy and the Family, 15, 88–101. Palloni, A. (2001). Increment-decrement life tables. In: S. H. Preston, P. Heuveline & M. Guillot (Eds), Demography: Measuring and Modeling Population Processes (pp. 256–272). Oxford, England and Malden, MA: Blackwell. Popenoe, D. (1988). Disturbing the nest. New York: Aldine De Gruyter. Popenoe, D. (1993). American family decline, 1990–1990: A review and appraisal. Journal of Marriage and Family, 55, 517–542. Popenoe, D. (1996). Life without father. New York: Martin Kessler Books. Poulain, M., Riandey, B., & Firdion, J. M. (1991). Enquˆete biographique et registre Belge de population: Une confrontation des donn´ees [Life History Biographies and the Belgian Population Register: A Comparison of the Data]. Population, 46, 65–88. Preston, S. H., Heuveline, P., & Guillot, M. (2001). Demography: Measuring and modeling population processes. Oxford, UK: Blackwell Publishers. Prinz, C. (1995). Cohabiting, married or single: Portraying, analyzing and modeling new living arrangements in the changing societies of Europe. Brookfield, VT: Ashgate Publishing. Smock, P. J. (2000). Cohabitation in the United States: An appraisal of research themes, findings, and implications. Annual Review of Sociology, 26, 1–20. Toulemon, L. (1997). Cohabitation is here to stay. Population: An English selection, 9, 11–56. Uhlenberg, P., & Eggebeen, D. (1986). The declining well-being of American adolescents. The Public Interest, 81, 25–38. van de Kaa, D.J., (1987). Europe’s second demographic transition. Population Bulletin, 42, 1. Washington, DC: Population Reference Bureau.

278

JEFFREY M. TIMBERLAKE AND PATRICK HEUVELINE

APPENDIX A Table A1 presents survey dates, female respondent upper age limits, and female and child sample sizes for the 17 FFS surveys analyzed in this chapter. Table A1. FFS Survey Dates, Respondent Upper Age Limits, and Sample Sizes. Survey Dates Country Austria Belgiuma Canada Czech Republic Finland France Germany Hungary Italy Latvia New Zealand Poland Slovenia Spain Sweden Switzerland United States a The

From 12/95 12/91 1/95 11/97 8/89 1/94 7/92 12/92 11/95 9/95 9/95 12/91 11/94 11/94 10/92 10/94 1/95

Female Age Range

n

To

Survey

Period of Analysis

Females

Children

4/96 12/92 12/95 12/97 1/90 5/94 7/92 12/93 1/96 10/95 11/95 12/91 11/95 11/95 5/93 6/95 10/95

20–55 20–42 15–55 15–45 22–53 20–51 20–40 18–42 20–50 18–50 20–60 18–50 15–46 18–50 22–45 20–50 15–45

18–45 20–37 15–45 15–40 20–45 18–45 20–36 18–37 18–43 17–45 18–45 18–45 15–41 18–45 20–38 18–43 15–40

4,535 3,231 4,050 1,730 4,140 2,936 5,990 3,590 4,794 2,688 2,901 4,210 2,788 4,021 3,272 3,876 10,711

7,039 3,771 5,452 2,363 6,220 4,525 5,923 4,933 5,495 3,902 5,847 7,016 4,005 5,065 4,686 5,440 14,824

Belgium survey only covers the Flanders region.

ABOUT THE AUTHORS

Carles Alsinet is Professor of Social Psychology in the Faculty of Pedagogy and Psychology at the University of Lleida, Spain. His primary research interests are on children’s rights and children’s well-being. Loretta E. Bass is an Assistant Professor of Sociology at the University of Oklahoma. She focuses her research on children and stratification issues, and completes research in West Africa and the U.S. She recently completed a book, Child Labor in Sub-Saharan Africa, (Lynne Rienner Publishers, 2004) which offers a window on the lives of Africa’s child workers drawing on research and demographic data from 43 countries. Dr. Bass’ research has appeared in Population Research and Policy Review, Political Behavior, Anthropology of Work Review, and the International Journal of Sociology and Social Policy. Michael F. C. Bourdillon was born in Africa and has spent most of his life in Zimbabwe. He is a social anthropologist, who has taught for over 25 years in the Department of Sociology, University of Zimbabwe. He has researched and published extensively on African religion. In recent years, his focus has turned to disadvantaged children in Zimbabwe. Apart from his academic work, he has long worked with an organization supporting street children in Harare. He has also cooperated with Save the Children Alliance, facilitating the establishment of a movement of working children in that country. ¨ Doris Buhler-Niederberger is Professor in Sociology at the University of Wuppertal, Germany. Several of her recent research projects have concerned childhood as a domain of professional, moral and political interest and images of childhood and children in public and professional debates. Her teaching and research interests are mainly focused on the sociology of private life and on private strategies of production and reproduction of social status and social order. Suellen Butler is currently the College Program Head of Urban Education (URBCC) and soon will be the coordinator of the Elementary Education in Multicultural Settings (ELEDM) program at Penn State Delaware County. 279

280

ABOUT THE AUTHORS

Dr. Butler’s contribution to this volume explores the activities and practices of the National School and Community Corp (NSCC), an AMERICORP schoolbased mentoring program in Philadelphia. Dr. Butler examines in what ways these school-based mentoring programs impact the childhood experiences of children and their schools. Steve Carlton-Ford is Associate Professor of Sociology at the University of Cincinnati, and an affiliate with the Department of Sociology’s Kunz Center for the Study of Work and Family. His research examines both the impact of war on children’s life chances and the effect of chronic childhood illness (particularly epilepsy) on family relationships and children’s well being. He currently edits Sociological Focus, the journal of the North Central Sociological Association. Ferran Casas is Senior Professor of Social Psychology in the Faculty of Education and Psychology at the University of Girona, Spain. He is Director of the Research Institute on Quality of Life. He is author of many books and articles on children’s rights. His main topics of research are well-being and quality of life, children’s rights and intergenerational relationships. Verna Chow has training in neuropsychology and is a researcher at the University of Calgary. Verna Chow’s and Dr. Hiller’s contribution to this volume stems from a mutual interest in second-generation immigrants and their adaptation to Canadian Society, which officially proclaims itself as multicultural. Laura Daniel received a FAPESP Award as a student at Sao Paulo State University (UNESP) – Marilia, for researching “Toys and Games: Childhood in the Parque das Nac¸o˜ es Favela,” which was supervised by Dr. Ethel Volfzon Kosminsky. She is currently a Social Sciences Master’s degree student at the same university in Brazil, researching children and gender. Fabio Ferrucci is an Associate Professor of Sociology of Culture and Sociology of Education at the Faculty of Human Sciences of the University of Molise (Italy). His research focuses on the family, social policy and non-profit sector. He is the author of several articles on childhood and family policies in Italy. Cristina Figuer holds a Master’s in Psychology and is currently a doctoral student in the Psychology and Quality of Life Program and researcher of the Institute on Quality of Life at the University of Girona, Spain. Kevin M. Fitzpatrick is a Professor of Sociology at the University of Birmingham, Birmingham, Alabama. His primary research focus is on health-compromising behavior among children and adolescents. In addition, he continues his work examining the role of environments and their impact on the mental health and well-being of homeless, youth, and other high-risk populations.

About the Authors

281

M`onica Gonz´alez holds a Master’s in Psychology and is currently a doctoral student in the Psychology and Quality of Life Program and researcher of the Institute on Quality of Life at the University of Girona, Spain. Daniela Grignoli is a Researcher at the Department of Economics, Management and Social Sciences of the University of Molise (Italy). She teaches Sociological Methodology and conducts research on children and new technologies. Mireia Gus´o holds a Master’s in Economics and is currently a researcher of the Institute on Quality of Life at the University of Girona, Spain. Patrick Heuveline is Assistant Professor in the Department of Sociology at the University of Chicago, and Research Associate at the Population Research Center, NORC and the University of Chicago. His research centers on the family as an adaptive institution and its key role in linking macro-level changes and individual behaviors. He is currently studying the consequences of mortality change in Cambodia and in high HIV-prevalence populations in Southern Africa. In addition, he is launching an international study of the effects of the relationship between the family and the State on youth well being across Western countries. Harry H. Hiller is Professor of Sociology at the University of Calgary. His specialization is dealing with macro-level questions about Canadian Society and he is the author of Canadian Society: A Macro Analysis (Prentice-Hall, numerous editions). Dr. Hiller’s and Verna Chow’s contribution to this volume stems from a mutual interest in second-generation immigrants and their adaptation to Canadian Society, which officially proclaims itself as multicultural. David A. Kinney received his Ph.D. in Sociology from Indiana UniversityBloomington and did post-doctoral work at the University of Chicago. He is currently Associate Professor of Sociology at Central Michigan University and an affiliate faculty member at the Center for the Ethnography of Everyday Life at the University of Michigan. In addition to being the current Co-Series Editor of the Sociological Studies of Children and Youth with Katherine Brown Rosier, his publications have appeared in Sociology of Education, Youth and Society, Personal Relationships During Adolescence (Sage), and New Directions for Child and Adolescent Development (Jossey-Bass). He is currently conducting ethnographic research with children and their parents in a study of how families manage work, home life, and children’s activity involvement in a fast-paced society. Ethel Volfzon Kosminsky, Professor of Sociology at Sao Paulo State University (UNESP) – Marilia, has been a recipient of research grants from the Brazilian organization CNPq and was a Fulbright grantee in the U.S. in 1995. Chair of the Graduate Program of Social Sciences at UNESP-Marilia from 2000 to 2004, she

282

ABOUT THE AUTHORS

currently leads the Center of Studies of Children and Adolescents at UNESPMarilia, and the Network for the Study of Latin American Children and Youth. Madeleine Leonard is a Reader in Sociology at the School of Sociology and Social Policy, Queen’s University, Belfast. Her research interests fall within the broad remit of the “new sociology” of childhood and she has conducted research with children on a wide range of topics including their experiences of poverty, their experiences of paid employment and their participation in domestic labor within the household. Her current research concerns Protestant and Catholic children growing up along one of the most contentious peace-lines in Belfast and the research examines children’s roles as political actors in Northern Irish society. Antonio Mancini is a Junior Researcher at the Department of Economics, Management and Social Sciences of the University of Molise (Italy). He is the author of several articles on children’s rights. He has also co-edited a book about the rights of the children. Hyunjoon Park is a Doctoral Candidate in Sociology at the University of Wisconsin-Madison. His research focuses on the process of transition to adulthood, particularly among young people in East Asia, across several dimensions including educational and occupational attainment. Currently, he is working on a dissertation project that compares the effects of family and school on educational achievement among 15-year olds in 30 countries using the PISA data. Recent publications include “Age and Self-Rated Health in Korea: A Research Note” (Social Forces, forthcoming) and “Racial/Ethnic Differences in Voluntary and Involuntary Job Mobility among Young Men” (with Gary Sandefur, Social Science Research, 2003). Bettina F. Piko, M.D., Ph.D., graduated from medical school in 1991, then started her career in the field of public health. In the meantime, she earned an M.A. degree in sociology and a Ph.D. in health psychology and behavioral sciences. Currently she is an associate professor of behavioral sciences at the University of Szeged Hungary, and her research activities embrace research topics from psychosocial youth development, substance use and problem behavior, up to psychosocial work environment, social support and societal stress. Samantha Punch is a Lecturer in Sociology in the Department of Applied Social Science at Stirling University. She recently completed a British Academy Postdoctoral Fellowship during which time she conducted a study of children’s experiences of sibling relationships and birth order in the U.K. Prior to this, she worked with Roger Fuller, Christine Hallett and Cathy Murray on the project “Young People and Welfare: Negotiating Pathways” which explored Scottish children’s problems and their coping strategies, as part of the ESRC’s Children

About the Authors

283

5–16 Programme. Her doctoral research included two years of ethnographic fieldwork on rural childhoods in Bolivia where she investigated the ways in which children and young people negotiate their autonomy at home, school, work and play. Marina Rago is a Junior Researcher at the Department of Economics, Management and Social Sciences of the University of Molise (Italy). She is currently involved in research projects on the implementation of children’s rights. Katherine Brown Rosier is currently Associate Professor of Sociology at Central Michigan University. She published Mothering Inner-City Children in 2000 with Rutgers University Press and is currently the Co-Series Editor of the Sociological Studies of Children and Youth with David Kinney. Other publications have appeared in The Journal of Contemporary Ethnography, Human Development, The Journal of Comparative Family Studies, and several other journals and edited volumes. While continuing to write on experiences of low-income AfricanAmerican children and families, she is also conducting research and writing a book with colleague Scott L. Feld on Louisiana’s Covenant Marriage. Carles Rostan is Professor of Developmental Psychology in the Faculty of Education and Psychology at the University of Girona, Spain and researcher of the Institute on Quality of Life. His primary research interests are on children’s development and children’s rights. Marta Sadurn´ı is Professor of Developmental Psychology in the Faculty of Education and Psychology at the University of Girona, Spain. She is researcher of the Institute on Quality of Life. Her primary research interests are on children’s development and children’s rights. Gary D. Sandefur is Professor of Sociology at the University of WisconsinMadison. His publications include Growing Up with a Single Parent (Harvard University Press, 1994) with Sara McLanahan, “What Happens after the High School Years among Young Persons with Disabilities,” Social Forces, 82 (2003), 803–832 with Thomas Wells and Dennis Hogan, and “Off to a Good Start? Postsecondary Education and Early Adult Life,” in Richard Settersten, Frank Furstenberg, and Ruben Rumbaut (Eds), On the Frontier of Adulthood: Theory, Research, and Public Policy, University of Chicago Press, forthcoming with Jennifer Eggerling-Boeck and Hyunjoon Park. He is currently working on quantitative and qualitative analyses of the transition to adulthood in the United States and other countries. Angelo Saporiti is Professor of Sociology at the Faculty of Economics of the University of Molise (Italy). Dr. Saporiti also teaches Social Ethics, and is the

284

ABOUT THE AUTHORS

author of books and articles on children’s rights. Angelo Saporiti is involved in various research international networks on childhood sociology and children’s rights. Jeffrey M. Timberlake is Assistant Professor of Sociology and Research Associate at the Kunz Center for the Study of Work and Family at the University of Cincinnati. He primarily studies the causes and consequences of urban inequality, particularly race-ethnic residential segregation. Current projects include analyzing data from the 1970 to 2001 Panel Study of Income Dynamics and the 1970 to 2000 U.S. Censuses to estimate racial inequality in children’s neighborhood socioeconomic status. In addition to his work with Patrick Heuveline on comparative family demography, he is also conducting several studies of raceethnic attitudes in America. Darlene Romania Wright is Assistant Professor of Sociology at BirminghamSouthern College. Her primary research interests pertain to adolescent healthcompromising behavior. Her current research is on the effects of social capital on violent behavior among secondary school students.