Race, poverty, history, adoption, and child abuse: Connections

Dorothy Roberts, Shattered Bonds: The Color of Child Welfare. New York: Basic Books, 2001. x + 349 pages. $26.00 cloth; $16.50 paper.

E. Wayne Carp, ed., Adoption in America: Historical Perspectives. Ann Arbor: University of Michigan Press, 2002. $57.50 cloth.

Introduction

Professor Dorothy Roberts, the Kirkland and Ellis Professor of Law at Northwestern University Law School, has written a compelling book on the inadequacies of the contemporary child welfare system, focusing on its differential impact on African-American and white children. She argues that the child welfare system has systematically dismantled the African-American family. Shattered Bonds: The Color of Child Welfare (2001) is utterly persuasive in documenting how the foster care system disproportionately and seriously affects and harms black families and children, including both the decision to remove children and the length of stay in foster care. In the past, Roberts has written compellingly about the role race plays in criminal law, reproduction, and other contexts.1 Nonetheless, I am not completely persuaded that the child protective system deliberately dismantles black families.2 Class, rather than race, still seems to me to be one of the dominant motivations for exposing children to the child welfare system, although, of course, class and race in this country are closely related.

In Adoption in America: Historical Perspectives (2002), Professor E. Wayne Carp, who teaches at Pacific Lutheran University, offers an edited collection of nine articles on the history of adoption, with contributions ranging from an examination of adoption in 19th-century literature to an analysis of the mid-20th-century effort by adoption agencies to create the perfect family.3 The scholarly interest of the contributors ranges from history to sociology to literature. Carp's goal is to identify and develop adoption history as a field unto itself, one that has broad implications for the definition of families and the relationship among parents, children, and the state (2002:7). Carp is uniquely situated to edit this anthology; in a previous book (1998), he eloquently analyzed the history of secrecy and confidentiality in adoption, as perhaps the only researcher to have obtained access to original adoption records.

Each book is an extremely ambitious undertaking; examined together, they provide compelling insights into our current and historical methods for treating adoption and abused and neglected children. Although the child protection system and adoption are often seen as separate-child protective services protects children from abuse and neglect, while adoption finds new families for children-they are, of course, integrally related. Adoption developed as the protection system came of age as a means for taking care of orphaned and abandoned children. Carp's anthology provides this history, and Roberts' book critiques the current status of the system.

Adoption developed as a way to care for white children; it has typically been less utilized in the African-American community for a variety of reasons (Solinger 1992; Berebitsky 2001). However, as Carp's anthology shows, adoption has been a means of socializing culturally disfavored children, removing them from their families of origin and placing them in middle-class homes, a practice not too different from what Roberts believes occurs today for black children. Indeed, even though adoption developed as a peculiarly American institution for myriad reasons, one of the primary impetuses for general adoption legislation was the growing number of children entering public care (Cahn, forthcoming [b]). We think of adoption in the context of infants, but it was not until after World War II that more infants than older children were adopted (Carp 2002), and the older children were frequently adopted after some contact with the child welfare system. Thus, the two systems have been intertwined for at least the past century and a half.

In this review, I provide a summary of the two books, and then place them in the context of the history of the legal treatment of abused and neglected children. Notwithstanding the disparate purposes and historical periods of the two books, they both address three broad themes: (1) the relationship among class, race, and child welfare; (2) the tensions between biological and functional families; and (3) the potential conflicts between parents' and children's interests.

The contemporary child protection system has its origins in the policies and practices discussed by several contributors to Carp's anthology. While the race of the children in 19th-century protective care was almost invariably white, they were from European cultures considered to be outside of the mainstream.4 Thus, the function of the abuse and neglect system-the assimilation of culturally disfavored groups-has been historically consistent (Lawes 2000). Moreover, adoption has served a variety of goals throughout this period. Although protecting children's interests has been one goal, adoption has also served a variety of adult interests as well (Cahn 2002; Ross 1999:1261-62, 1264-77), ranging from finding children as laborers for a family to attempts to create the "perfect" family.

I. The Books

Although both books are concerned with child welfare, Shattered Bonds focuses on the relationship between the contemporary system and race, while Adoption in America provides a general history of child welfare, with a focus on adoption.

A. Adoption in America: Child Welfare History

The contributors to Carp's anthology document various aspects of the history of adoption in America.5 Each essay focuses on some aspect of the history of adoption, ranging from early18th-century orphanages to late-20th-century adoption narratives. In his Introduction, Carp provides a comprehensive history of adoption. One theme running through many of the chapters is that the historical vision of adoption cannot be separated from class, race, and religion.6 A second theme concerns the tension between blood and adoptive ties.

a. Race and Class in Adoption History

Turning to the first theme, as Linda Gordon has so persuasively shown, throughout the 19th century, the child savers were influenced by their own class situation. In her study of three Boston child welfare agencies, Gordon found that immigrant children were overrepresented in the caseload in comparison to their proportion of the Boston child population, although not in comparison to their proportion among the poor (Gordon 1988:9). In crafting the strategy of the Children's Aid Society (CAS) in late-19th-century New York City, Charles Loring Brace deliberately sought to sever the familial relationships between immigrant Catholic children and their parents.7 Throughout the 1850s, the annual reports of the Children's Aid Society described its clients as "[falling] short of being fully human ... ascribing a feral or beastly nature to the poor" (Bellingham 1984:32-33). The CAS did not want to return the children they had rescued to their "vile" parents (Carp 2002). Class has served as a critical marker in the development of adoption law, but it has also been intertwined with other issues of biology, heredity, and "matching" between adoptive parents and adoptees.

In the first chapter of the anthology, Susan Porter (Carp 2002:43) analyzes early adoptions, those before the mid-19th century, by examining the practices of orphanages that attempted to place children. Adoption was a complex institution, she concludes. The women who organized the early orphanages believed that they were helping children by providing them with middleclass families (Carp 2002:47). The managers of the orphanages anticipated that adopted children would not be treated like indentured servants, but would achieve emotional closeness within their new families. Instead, adoptive families remained somewhat ambivalent about whether the adopted children really were new members of their families or were apprentices; thus, the orphanage managers became somewhat frustrated. Moreover, in accordance with 19th-century values, managers came to believe that children should be returned to their relatives (Carp 2002:54, n20; Cahn, forthcoming [b]), and the birth family often viewed placing children in the asylums as only temporary (Carp 2002:46). Early efforts to create non-consanguineous families were quite difficult; the creation of adoption as a separate status from indenture cannot be crystallized at any one moment.

Porter's conclusions are supported by studies of other child-- saving organizations. In 1849, the state of Massachusetts enacted Articles of Incorporation for the Worcester Children's Friend Society. The Society was chartered as a child-saving organization, and soon it established a home for poor children (Lawes 2000:122). Rather than operating an asylum for children, however, the Society sought foster families willing to care for the many children who needed a new home (Lawes 2000:123, 143-44).8 Pursuant to its legislative charter, the Society was responsible for placing the children with "virtuous and respectable citizens." Many of the Society's founding members volunteered to take in these children; indeed, not only did Ann Buffman Earle, a board member of the Society, agree to provide a home for six-year-old John Tyler, she convinced her sister to become a foster parent for nine-year-old Emma Tyler (Lawes 2000:141-42). For members of this society, finding other families who would provide the appropriate nurture and support was difficult, even though the Children's Friend Society, in its first five years, placed 62 children who had a strong expectation of adoption by their foster families (Lawes 2000:145). Nonetheless, when the managers of orphanages sought to find homes for the children, they were frequently told that the adopted children would never be treated in the same way as biological children. Indeed, " `the remark so often heard at the home, from persons applying for a child, [is] I do not expect to love or treat this boy or this girl as I do my own children"' (Lawes 2000:142).

Indeed, many families who accepted children from the aid societies in this period explicitly indicated their interest in labor help.9 People who accepted children from the "orphan trains," the trains that brought New York City orphans to western towns for placing-out, were frequently looking for cheap labor, and there were many complaints that the foster children were being overworked (Gordon 1999:9-10; Bellingham 1984). Placing-out children did serve to provide a permanent home for some poor children, but it also guaranteed permanent labor for their adoptive parents. During the middle part of the 19th century, children were wanted for their ability to perform household labor. Even at the turn of the century, more than 15% of children between the ages of 10 and 15 were working, and this number does not include children who helped their parents on farms, who worked in sweatshops, or who were working even though they were younger than ten (Zelizer 1985:56). Biological children were expected to perform a fair amount of labor in their families, but, with the decline in indentured servitude, some families sought foster children for this purpose. In some situations, of course, the children were accepted as family members, and even changed their names (Lawes 2000:145). The gap between blood and adoption, however, remained wide throughout this period (Lawes, 2000:143).

Novels from the period, as discussed by Carol Singley in her chapter (Carp 2002:100-101), reflected the general cultural atmosphere, emphasizing the importance of bringing up immigrant children in good American homes. As Paula Pfeffer (Carp 2002) also observes, so strong was the orientation toward providing good American Protestant homes, that Catholic and Jewish adoption agencies developed in opposition to efforts by Protestant organizations, such as the Children's Aid Society, to indoctrinate their children.10 Other novels in this period stressed the tension between relationships by blood and by adoption. Singley observes that, prior to the Civil War, the literary theme of adoption appeared in the service of religious transformation; ofter the Civil War, adoptees became commodified objects (Carp 2002:109-11). The adoptive parents in fiction viewed adoption ties as weaker than blood ties (Carp 2002:112).

In another discussion of adoption narratives in literature, Barbara Melosh links these stories to larger cultural themes (Carp 2002). Both she and Singley believe that adoption literature is so significant in popular culture because its themes transcend the particular issues of any adoption triad and resonate with people's deeper issues of identity (Carp 2002:374). Melosh discusses these identity issues for individuals' constructions of selves, or in autobiographies, and Singley discusses the identity issues in the context of 19th-century national identity. Notwithstanding its resonance, however, adoption stories serve to mark the difference between adoptive kinship and biological kinship, differences framed by absence and stigma (Carp 2002:344-45).

In her chapter in Adoption in America, Patricia Hart relates that early-20th-century child-savers similarly hoped to reform children, removing them from their families of origin so they could be socialized into appropriate citizens. She observes that the motto, "It is better to save a child than retrain a criminal," expressed the sentiment that adoption could prevent a child from leading a life of delinquency (Carp 2002:185). In 1911, the Washington Children's Home Society explained that a legitimate reason for a birth parent to place a child for adoption was to -protect society and the state by the transformation of a prospective vagrant and criminal into a noble [a code word for native-born and white], high-minded, helpful citizen"' (Carp 2002:187, n16). A 1922 article in the Columbia Law Review summarized the benefit of adoption, "[I] n the majority of cases [taking] a child from a home or poverty or a charitable institution and placing that child in an environment tending to his physical, mental and moral uplift and betterment" (Brosnan 1922:332, 341). In 1938, Grace Abbott, the former Director of the U.S. Children's Bureau, noted that "[t]he practice of taking children from their parents solely on the ground of poverty is rapidly disappearing" (167). Her optimism was, perhaps, prompted in part by the recent enactment of Aid to Dependent Children.

b. Adoptive and Biological Kinship

Throughout several chapters (Carp 2002) are questions about the historical relationship between biological and adoptive families. Should adoptive families look like other families? How should adoptees and adoptive families be chosen? Can adoptees overcome their biological heritage? As 19th-century courts and legislatures struggled to assimilate adoptive families into a legal system based on respect for the biological family, they reified the biological family as the legal and social norm for all other families. Early adoption law and practices sought to assimilate functional families into existing norms, in an effort to provide functional families with the same privileges as biological families.

During the middle of the 20th century, as adoption agencies and the practice of social work sought to become more established,11 they became increasingly involved in the adoption process. Adoptive families and potential adoptees were carefully scrutinized in an attempt to create a "normal" family.12 Social workers sought to match children and families based on intelligence, religion, and race. During the mid-20th century, many parents did not tell their children that they were adopted, lest the family be seen as different, and worse, from other families (Carp 2002:126). By the middle of the 20th century, social workers believed that the happiness of adoptive parents and of adopted children depended on this matching. Indeed, matching became one of the principal strategies for lessening the stigma of adoption by attempting to replicate the family that the adoptive couple would have had, absent adoption (Cahn 1994; Samuels 2001). This included attempts to match people of the same ethnic origins. Writing in 1919, in the early days of matching, Dr. Slingerland ([1918] 1974) observed that it was "desirable in fitting children to applications, to select such as resemble one or both of the foster parents. . . . It is also worth while to avoid mixing too diverse types or nationalities."13

Many vestiges of this matching strategy still exist (Bartholet 1999a:72). The child receives a new birth certificate, listing the adoptive mother as the real mother, literally obliterating the existence of the birth mother. Older people are discouraged from adopting newborns to preserve a "normal" familial age. White parents are discouraged from adopting black children (Shanley 2001). Some state statutes explicitly direct that children will be matched with families of the same religion, or they permit agencies to consider religious, ethnic, and racial heritages.14 Such strategies, on one hand, reinforce the primacy of the biological family, suggesting that families that look different are different; on the other hand, they respect the needs of the child.15

The continuing cultural preference for biologically based families can be seen in the first general study of public attitudes toward adoption. In a 1997 survey of more than 1,500 adults, 90% of the participants had a positive opinion of adoption, and 95% generally supported it. Nonetheless, when it came to an examination of the adoptive family, respondents were somewhat more cautious. Half of the respondents believed that, although adopting a child was better than being unable to bear a child because of the woman's or man's infertility, it was not quite as good as having a biological child (Wetzstein 1997).16 Only twothirds believed that it was highly likely that an adoptee would love her adoptive parents as much as her biological parents. There is, then, continuing ambivalence with respect to families formed through adoption rather than through biology, a belief that blood ties are stronger and more desirable than adoptive ties.

Yet the effort to promote general cultural acceptance of adoption can be seen as early as 1907, when The Delineator, a woman's magazine, in a series of articles promoting adoption of poor orphaned children, publicized the availability of children for adoption through its "Child-Rescue" campaign.17 Berebitsky, who discusses this campaign in her contribution to Carp's (2002) anthology, claims that it was very successful-at least 2,000 children were placed during the campaign's three and one-half year duration-it produced many adoptive families and helped to decrease the stigma attached to adoptive families (Carp 2001).18 The articles emphasized women's ultimate role as mothers and also sought to overcome their fears of adopted children's "bad" heredity. Adopting the poor immigrant children would fulfill women's gendered duty, but it would also save society through the resocialization of these children. Adoption, as depicted in The Delineator's campaign, was again class-based, with American middle-class women rescuing outcast orphans.

As yet another means of fostering acceptance of adoptive families, mid-20th-century adoption agencies shaped adoptive families to conform to the dominant family forms. Gill (in Carp 2002) argues that the agencies defined child welfare in terms of creating perfect families. Instead of placing disabled and minority children in new homes, they were disadvantaged by a policy that narrowly defined the perfect family by race and ability. Among Gill's fascinating findings are a 1950 California social welfare regulation directing that the child's personality, intelligence, cultural background, and coloring be considered relevant to finding the appropriate adoptive family (Carp 2002:250, n36). Obviously, there were no transracial adoptions in this period. Moreover, African-American potential adoptive parents had even more difficulty adopting than did white parents, notwithstanding the numbers of black children waiting to be adopted (Carp 2002:251, n40 and n41). The meaning of "child welfare" depended on social workers' cultural understandings of appropriate family forms.

Like Gill, other contributors to Carp's anthology note that perfect families also depended on appropriate gender roles. Adoptive mothers who worked outside the home, for example, did not conform to the middle-class-breadwinner-and-- dependent-wife household sought for adoptive parents (Carp 2002:194).19 The tension between what is best for any particular child and the cultural parameters of good families appears in these agency practices, just as Roberts (2001) alleges it does in the contemporary child welfare system, where remaining in a biological family may mean that a child stays with an undeserving mother.

In their joint contribution to the anthology, Carp and Anna Leon-Guerrero (Carp 2002:278) argue that World War II marked a transformation in adoption policy and practices. Based on their unique access to the adoption case records of the Children's Home Society of Washington and extensive statistical analysis of the attributes of the adoption triad members contained in those records, they conclude that innovations in social work and changes in the demographic composition of triad members resuited in dramatic changes to adoption. During this period, the age of birth parents decreased, while their marital status changed from married to single parents. For example, the age of birth mothers between 1900 and 1939 ranged from 24 to 27; between 1940 and 1973, the age ranged from 19 to 24 (Carp 2002: Table 8.1). Adoptive parents also developed a preference for younger children, with approximately 50% of parents requesting children over age three prior to World War II. Over the next 30 years, the age of children requested changed, and by the 1970s, virtually all adoptive parents preferred infants (Carp 2002:299). Carp and Leon-Guerrero speculate that after World War II adoptive parents were also less concerned about inherited genetic traits and infants' vulnerability to death (Carp 2002:278)20 believing instead in the ability of parental love and nurture to overcome any potential disability (293). The sentimentalization of children, they argue, was a gradual process that culminated post-World War II (298).21

B. Shattered Bonds: Child Welfare Today

The goal of Shattered Bonds is, quite bluntly, "to call the child welfare system what it is: a state-run program that disrupts, restructures, and polices black families" (Roberts 2001:viii). Roberts divides the book into three parts: in the first part she documents the racist impact of the child protection system; in the second she examines the changing politics of child welfare, which has resulted in the increased severing of the parent-child relationship; and in the final section she provides a theoretical perspective on the system's harm to the black family and to AfricanAmericans as a group.

In the first major section, Roberts provides extensive data on the overrepresentation of blacks in the system that tracks child abuse and neglect. She explains that the racism in the system has changed forms over the past century. In the early 20th century, until the mid-1940s, black children were essentially excluded from child welfare services (Roberts 2001:7). Indeed, black families were excluded from many of the social welfare systems available to help families in need (Skocpol 1995). This situation has changed; a 1961 study by the U.S. Department of Health and Human Services, Administration on Children, Youth, and Families (HSS) (2001:3) found that minorities were disproportionately likely to be receiving public services for child welfare. Today, at the beginning of the 21st century, Roberts observes, although only 17% of the nation's children are African Americans, they represent 42% of the children in foster care (2001:8). She carefully explores whether poverty or race is the reason for this disproportionality. Roberts uses studies showing that, even when controlling for poverty, family structure, parental employment, and location, black children are, first, more likely to be labeled as abused than white children (2001:49), second, more likely to be placed in foster care (52), and, third, more likely to remain for longer periods in foster care (23). Once they are involved in the child protective services system, she argues, AfricanAmerican children receive poorer quality services than do white children (20). In her view, racism, not poverty, provides the explanation.

Even if the racial disparity could be explained by higher Black poverty rates and not intentional discrimination, this would not negate the racist impact of the system or the racist reasons for its inequities. Racism often involves but does not require prejudice against Blacks. . . . [R]acism is a system of white privilege....

My answer to critics who demand to see evidence of ill will against Black families, then, is that racial motivation is not necessary to show that the system discriminates. We should not ignore, though, the considerable evidence that race and not poverty alone affects decision making at every step of the child protection process (Roberts 2001:95).

After showing the racial disparities that exist throughout the child protective services system, Roberts discusses recent federal and state efforts that exacerbate these disparities. She elegantly interweaves information about federal legislation concerning child abuse and welfare reform, as well as that concerning more stringent criminal laws that result in the increasing incarceration of African-American parents, arguing that this legislation will not strengthen the black family. The Adoption and Safe Families Act (ASFA) (1997), the most recent major federal legislation concerning the abuse and neglect prevention system, presumes to accept that adoption and foster care will provide children with the safe families they need (Roberts 2001:114). Although the ASFA reiterates previous law in requiring states to make reasonable efforts to preserve and reunify existing families,22 it places new emphasis on permanency planning and adoption. The primary purpose of ASFA was to increase the number of children adopted from the foster care system; indeed, the bill was titled the "Adoption Promotion Act" (1997) when it was passed in the House of Representatives. Rather than the previous focus on pursuing reunification before adoption, the new legislation allows for simultaneous pursuit of "reasonable efforts to place a child for adoption or with a legal guardian [and] reasonable efforts [to reunify families]." 23 Moreover, if reunification efforts would conflict with a "permanency plan," then the plan should take priority.24 The legislation also provides incentive payments to states to increase the number of children adopted out of foster care.

Spending on child welfare services, Roberts observes, is 12% of the spending on foster care; child welfare agencies are thus constrained in their spending on family preservation services (2001:142). She argues that the rush to terminate parental rights in ASFA shortchanges biological parents (150-52), and creates "legal orphans" who will futilely wait to be adopted (157-59). Undoubtedly, although ASFA has resulted in the adoption of increasing numbers of children, there remains a significant backlog of children available for adoption waiting in foster care limbo.25

Welfare reform presents comparable problems, and is based, Roberts argues, on the same images of "unfit" or "undeserving" mothers. Both aspects of the child welfare system-aid and prevention of abuse and neglect-are moving toward privatization (Roberts 2001:137). Incentives for marriage and child support enforcement, as well as work requirements, parallel incentives for adoption. Like other scholars in this area, Roberts ties the two systems together, arguing that providing insufficient aid to poor families results in child protective services claims that abuse and neglect are not due to poverty, but to deficiencies in the parents (2001:200; Ross 1999; Ross & Cahn 2000).

The third strand in the family disruption story is the criminal justice system, where blacks are also disproportionately represented (Roberts 2001:200-201). Roberts notes the large overlap between children in foster care and those in the juvenile justice system (2001:205-6), as well as the impact on the family of incarcerated parents. Given racial bias in the juvenile justice system, more black children are sent "into state custody-the likely result of adoption and welfare reform, as well" (220).

In the final section of the book, Roberts indicts the child welfare system, not only for its disproportionate impact on black families but also for its group-based harm to blacks. The child welfare system intervenes in, and then dissolves, black families, thereby reinforcing the second-class status of blacks in American society (2001:225). She relies on the work of Charles Lawrence (1987), who argues that a showing of discrimination does not require proof of an actual discriminatory purpose (Roberts 2001:244-45). Lawrence uses a "cultural meaning" test to determine whether a government policy results from "unconscious racism" (1987:326). Pursuant to his test, a policy is discriminatory if it has a "racially stigmatizing meaning," that is, if it is understood as portraying blacks as inferior (Lawrence 1987:386). The child welfare system conveys just such a message, argues Roberts.

Although Roberts articulates the claim that the administration of child welfare services inflicts harm on African-American children, she does not allege that there is an essential black cultural identity per se that is damaged; instead, she explains that the child welfare system prevents black families from developing their own identities (Roberts 2001:254). She analogizes the removal of black children from their families of origin to the removal of Native American children from reservations and to the removal of aboriginal children from their families in Australia (2001:248-252). In each case, she argues, there has been a deliberate effort to place children in white families to sever them from their cultural origins.

Her solutions center on placing control over the child welfare system within the black community, and refocusing the system on supporting families, not on failing individual parents. Accordingly, she briefly makes the case for improved public welfare for families, including an increase in the minimum wage, a national health insurance program, available and high-quality child care, and paid parental leaves (Roberts 2001:268). She also favors increasing client participation in the child welfare system. Many of her suggestions are similar to those advocated by a wide variety of poverty and child welfare experts, underscoring the broad agreement across political and philosophical perspectives on how to fix the system.

II. Context

The two most significant challenges to Roberts' conclusions are, first, that it is poverty and other risk factors, not race per se, that accounts for the disproportionate representation of AfricanAmericans in the system, and, second, that regardless of the problems with the system, we should be concerned with child welfare rather than with protecting parents' rights to raise their children. With respect to each of these ongoing scholarly and practical debates, Shattered Bonds develops an articulate stance.

A. The Poverty/Race Explanation for Disproportionality

Although virtually all researchers agree that African-American children are overrepresented in the child protective services population, there are three theories that attempt to explain this overrepresentation. The first, as Roberts eloquently explores in Shattered Bonds, is race-based; the second is class-based; and the third justifies the numbers of African-American children in the child protective services system as proportionate to their actual needs (Hill 2001).

a. The Poverty Explanation

Poverty and child abuse are closely, albeit unfortunately, related. The rate of reported abuse and neglect is 22 times higher for children whose yearly family income is less than $15,000 than it is for children whose family income is greater than $30,000.26 In a study of welfare and child abuse completed prior to full implementation of the Temporary Assistance to Needy Families (TANF) statute, the researchers found that the families of almost 60% of the children entering foster care had a recent involvement with the public welfare system, although less than 3% of children in the public welfare system received out-of-home care (Hutson 2001:8).

Public policy analysts have begun to study the relationship between welfare reform and child abuse (Paxson & Waldfogel 1999, studies also cited in Hutson 2001). In their extensive study of child abuse and neglect from 1990-1998, based on state-level information from the National Center for Child Abuse and Neglect, Christina Paxson and Jane Waldfogel examined the impact of family structure on rates of child maltreatment. They found that higher rates of poverty resulted in higher rates of substantiated reports of abuse and neglect (1999:13). They also found that an increase in welfare benefits results in reducing cases of neglect and of foster care placements (16). In their research, although TANF may not have resulted in increased cases of child maltreatment, TANF and family caps did result in an increased number of children in out-of-home care (23). The poverty rate among African-American households, whether married or singleparent homes, is almost double that of the rate among white households.27 The victimization rate for African-American children is 25.2 per 1,000 children of the same race, while the comparable rate for white children is 10.6; overall, in 1999, there were 419,858 white child victims of maltreatment compared to 204,193 black child victims.28 A recent study in Denver of child abuse found that children in single-headed African-American households were more likely to be referred because of abuse than were white children who had also been abused while living in two-parent households (Jenny et al. 1999).29 Physicians missed child abuse in white children at a rate of about 40%; for black children, it was 20%. There were comparable rates for single-- headed versus two-parent families.

I think these statistics show that the abuse and neglect prevention system is not administered even-handedly, and they also show that there is a high correlation of poor families and the abuse and neglect prevention system. More than one-third of the children in New York City's foster care system also receive public welfare. While African-American children are disproportionately likely to receive child protective services and to be placed in foster care, it is unclear whether this is due primarily to racial animus or poverty or to other risk factors.30

b. Overrepresentation Accurately Represents Risks

In one of the few studies to examine the relationship between African-American children and the child welfare system, Richard Barth and his colleagues, on one hand, concede that African-American children are disproportionately involved in the child welfare system. On the other hand, they concluded that "when many factors are considered, African-American children are not overserved or overinvolved in the child welfare system" (Barth et al. 2001:61). In their careful review of the data from the National Child Abuse and Neglect Data System, they also found that "there is no compelling finding that race alone contributes substantially to child welfare decision-making" (Barth et al. 2001). Instead, they suggest that African-American children are at higher risk for needing child protective services, in part because of the higher rate of incarceration and mortality among their parents. Other risk factors, such as poverty, large family size, urbanization, and receiving welfare are also correlated with child protective services involvement.31 Moreover, they conclude that the provision of child protective services is not necessarily detrimental to the children (Barth 2001:xii); receipt of child protective services appears to decrease mortality as well as juvenile justice system involvement(113).

In a subsequent paper, Barth (2001:23) suggested that a "multiplicative model [which] posits that there are small to medium increases in the disproportionality by population experienced by African American children as they move through the child welfare system which, finally, result in a substantial difference in their representation in child welfare."

However, these conclusions may be challenged. Based on the National Incidence Study of Child Abuse and Neglect, there appears to be no difference in risk of harm for black and white children. Indeed, white children appear to be at a slightly higher risk of physical abuse, sexual abuse, and physical neglect (Sedlak & Schultz, 2001:13). Additionally, the National Incidence Study found that when risk factors such as family size, single-parent head, and poverty are taken into account, black children have a lower risk of maltreatment (2001:19).32 Even though AfricanAmerican children involved with child protective services may have more problems than white or Hispanic children, when families with the same characteristics and problems are compared, black children are more likely than white children to be placed in foster care (U.S. Department of Health and Human Services [HSS] 1997:21). According to Roberts, this disproportionality results from racism, regardless of whether there is direct proof of discriminatory intent.

The debate over whether it is race discrimination or poverty that primarily causes the overrepresentation of black children in the system, while important, is perhaps less significant than an analysis of what to do about the child abuse and neglect prevention system. The focus of the system on rescuing children, rather than on preventing their abuse or neglect in the first place, results in too many children not living in their families of origin.

There are many kinds of interventions before a child is removed that may be effective in preventing his or her removal: parenting classes, home visiting, helping parents find housing and jobs, coordinating public welfare services and domestic violence interventions with the child welfare system, and providing more intensive substance abuse programs. Even after substantiation of abuse or neglect, studies have found that the majority of children in out-of-home care could safely live at home (Guggenheim 2000:1724).

Leroy Pelton has suggested restructuring the child protective services agency to focus on these issues, rather than on investigative- or law-enforcement processes (1993, 1997). Over the past 30 years, the child protective services system has focused on removal at the expense of preservation; the number of children receiving in-home services declined by 60% from 1977 to 1994 (U.S. HSS 1997:2).3 Such a restructuring might make clients more comfortable seeking and accepting preservation services if they know they are getting help rather than confronting a high risk of child removal. Several states have implemented screening programs in which the child welfare agency investigates only the most severe cases of alleged abuse and neglect, while other cases are referred for family assessment and support.34 A broader vision of child welfare services would involve support for children's existing needy families and less focus on punishing those families (Guggenheim 2000:1746-47).

B. Child Protection, Not Parents' Rights

The arguments for protecting children regardless of the paren ts' situations are presented most sharply, perhaps, by Elizabeth Bartholet. In her book, Nobody's Children, she does not dispute that "racial and social injustice [are] at the core of child abuse and neglect (2000:1999, 2000)."35 Nonetheless, she advocates that children be removed and placed for adoption more quickly because potential or actual harm to them, not protection of their parents, should be at the core of an abuse and neglect system. She critiques the "blood bias" of the current system, which strives to keep children with their parents or within their kinship group, and argues that parenting is defined by "social," not blood-based, bonds (Bartholet 1999:81-82). Bartholet does agree with Roberts that it is important to end the cycle in which parents, who are victims of this system, turn their children into victims.36 They disagree quite vehemently, however, as to how to end the cycle. Bartholet argues that it is important to move the children into other homes where they will receive the nurturing they need, rather than leaving them with their biological parents (2000:1999, 2000). In contrast, Roberts argues for more intensive resources directed at keeping children in their families. She advocates better public support for children and community-based and guided interventions.

This debate echoes other issues involving parents and children, and the appropriate structure for recognizing children's rights. Parents are constitutionally entitled to raise a child in the manner that they choose. In Meyer v. Nebraska, the Court held that the right of liberty "denotes not merely freedom from bodily restraint but also the right of the individual to contract, to engage in any of the common occupations of life . . . to marry, establish a home and bring up children." In Pierce v. Society of Sisters (1925), the right is phrased as harking back to Meyer. "The fundamental theory of liberty upon which all governments in this Union repose excludes any general power of the State to standardize its children.... The child is not the mere creature of the state; those who nurture him and direct his destiny have the right, coupled with the high duty, to recognize and prepare him for additional obligations." In Farrington v. Tokushige (1927), the Court held that restrictions on teaching foreign languages in Hawaii deprived "Japanese parents" of the right to direct their children's education.37 In Wisconsin v. Yoder, the Court observed that parental control over their children's religious and educational upbringing has "a high place in our society," and the state's interest in education must be balanced against "the traditional interest of parents with respect to the religious upbringing of their children." In Troxel v. Granville (2000), the Supreme Court reiterated that parents have a basic right to raise their children, and that the decisions of fit parents should receive great deference. Each of these decisions occurred within the context of a nuclear family; however, when an unmarried father challenged an intact nuclear family, he lost (Michael H. v. Gerald D.).

Parents' basic rights become attenuated as soon as the fitness of the parents) becomes questionable (Wyman v. James [1971]; tenBroek 1964, 1965); child abuse and neglect statutes are premised on this concept. Indeed, while courts pay deference to the notion of parental control, the state can remove children from their parents for abuse and neglect, it can require some form of schooling, and it can establish a minimum work age for children (Ross 1996:1571, 1586).38 Moreover, the Supreme Court has recognized that children have some basic minimal rights as well, which they can assert on their own behalf.

Scholars have debated the issues involving children's rights when they conflict with their parents. The Court has largely reinforced the notion that the traditional family unit provides adequate constitutional protection for children. Particularly in the substantive due process context, the Court has tended to equate children's interests with those of their parents and to protect children derivatively, through such doctrines as parental autonomy and familial privacy.39 When it comes to children's rights to receive adequate services to prevent abuse and neglect, the Court has generally reinforced the state's decisionmaking process rather than children's rights (Deshaney v. Winnebago [1987]; Sister v. Artist M. [1992]).40

Nevertheless, intervention in the family remains based not only on the form of the family but also on class. As Jacobus tenBroek originally pointed out almost 40 years ago,

[W]e have two systems of family law .... One is public, the other private. One deals with expenditure and conservation of public funds and is heavily political and measurably penal. The other deals with the distribution of family funds, focuses on the rights and responsibilities of family members, and is civil, nonpolitical and less penal. One is for underprivileged and deprived families; the other for the more comfortable and fortunate. (quoted in Handler)41

This two-tiered model pervades every aspect of family law. Parental rights depend on the class of the family.42 Wealthier families have always received more protection for their familial-based decisionmaking, as the very history of public welfare to children shows (Cahn 1997:965; Ross & Cahn 1999; Williams 1992:719).43

The interdependence of class and parental rights is illustrated by the story of Wyman v. James, decided by the Supreme Court in 1971. Ms. Wyman, a public welfare recipient, refused to allow her caseworker to visit her home. She told her caseworker that she would provide any information that was relevant to her continued receipt of welfare, but that the caseworker could not make a home visit. At the time, New York state law required home visits to public welfare recipients once every three months, and the various purposes were to verify information concerning eligibility for welfare, provide professional counseling, and prevent welfare fraud. Additionally, New York law specified that children would only be eligible for aid "if his home situation is one in which his physical, mental and moral well-being will be safeguarded and his religious faith preserved and protected" (Wyman [1971] at 312; Colb 1998:1642, 1720-23). Although a three-judge district court struck down the home visit requirement, the Supreme Court reversed that decision. The focus of the opinion was on distinguishing a true Fourth Amendment search; the "visitation" at issue here was not forced, compelled, or backed up by criminal penalties. There was, not surprisingly, a strong dissent by Justices Marshall and Brennan, which explicitly confronted the class implications of the case. They explained:

[I]t is argued that the home visit is justified to protect dependent children from "abuse" and "exploitation." These are heinous crimes, but they are not confined to indigent households. Would the majority sanction, in the absence of probable cause, compulsory visits to all American homes for the purpose of discovering child abuse? Or is this Court prepared to hold as a matter of constitutional law that a mother, merely because she is poor, is substantially more likely to injure or exploit her children? (Wyman [1971], at 341-42)

The dissenters also noted that the home visit was justified as ensuring adequate information for confirming the family's eligibility for public welfare; ironically, the federal regulations specified that the parent should serve as the primary source of that information.44 The intrusiveness of the home visit and the alleged necessity of independent verification indicate that the mother is not to be trusted.

Thus, one response to claims of an excessive focus on parents' rights to keep their children is to point out the class-based nature of the family integrity doctrine. In responding to the "children's rights" perspective, Roberts provides another view; she argues that it is in fact a violation of children's rights to remove them from their families of origin. For example, federal law favoring adoption "does not further the interests of most children in foster care" (Roberts 2001:257) who would be better served by remaining with their parents. Instead, she identifies children's rights as part of a larger struggle against oppression, so that all children will be valued. She does not deny that child abuse and neglect are bad for children, but her solutions are not to remove children from their parents, but to provide support so that parents can raise their own children. Rather than see children in isolation from their families and communities, she wants to place them in context. Nevertheless, it is interesting that she does not believe that kinship care, or preferential placement with the children's relatives over placement with a stranger, provides an answer because it still requires parents to turn over custody of their children to the state prior to the placement. (2001:261-62). Her emphasis is on providing material services to the family so that children do not have to be removed. She does acknowledge that black children should be protected from neglect and abuse, and sometimes even removed from their homes, but not through their involvement with a racially biased system (255).

Both the foster care system and TANF are premised on providing support for the child, in whatever familial unit he or she is currently living, rather than on providing support to the familial unit (Ross & Cahn 1999). The willingness to focus on the individual child reflects much broader social, legal, and philosophical notions about the child as a future deserving citizen of the state, who is valuable regardless of the suitability or desirability of his or her parents. (Prince v. Massachusetts [1944]; Zelizer 1985; Minow 1990:267-311).

One of the many shortcomings of the child abuse prevention system is its failure to look at other family members. Roberts advocates a community approach to handling child abuse and neglect, with services generally available to all members of the community. Additional solutions expand upon this contextual approach. A recent decision about New York's child protective services system found that the system was disproportionately removing children from battered women; the women were being blamed for their victimization, and their children were placed in foster care (Nicholson v. Williams [2002]; Ross 2002). The higher rate of incarceration among African-Americans (Butler 1997) means that parenting issues are particularly difficult; changing how the child protective services deals with incarcerated parents would also help racial disparities in the child welfare system (Cahn, forthcoming [a]; Hirsch et al. 2002). While more than 64% of the mothers in prison had lived with their children prior to incarceration, fewer than half of all mothers in prison had ever had personal visits with their children (Mumola 2002:5). When men go to prison, their children are most likely to live with their mother; when women go to prison, the children are most likely to live with other relatives or to be placed in foster care.45 Women are decontextualized, treated as prisoners, without acknowledging the "framework that renders invisible the interrelated webs of inequality: the woman's role within the family, her responsibility for children, [and] her economic circumstances" (Hirsch et al. 2002:230).

Conclusion

The history of adoption provides a prism through which to examine the creation of the parent-child relationship outside of the traditional blood-based family form.46 It provides insight into, and challenges for, contemporary debates on adoption as well as on the utility of applying the biological family as a template to other family forms. Reexamining the history of adoption in this country provides new perspectives on contemporary debates over adoption, such as the necessity for biological parental consent, or the public nature of adoption records47-as well as the contemporary focus of the abuse and neglect system.48 The different attitudes about blood relationships-that they are bad when the parents are poor and African-Americans, but good when the parents are middle-class and white Americans-is a clear theme in both books. What emerges is the sense that individuals, families, and communities are all profoundly affected through the actions of child protective services and the system of adoption and that a family-based perspective for all facets of these systems are important (Teitelbaum 1996). Removing children from their homes has implications for them as individuals and as members of a family; examining family functioning, rather than individual performance, provides the appropriate lens. Thus, offering support to parents helps children; improving education for children helps them as family members. Children and their parents need not be seen as autonomous and potentially conflicting rights-bearers in order to respect the interests of both in their shared relationship and to protect them from harm.49