Not in Families Like Us: The Social Construction of Child Abuse in America

Paper Delivered to: Emerging Theories/Merging Practices in Gender Studies, The Ninth National Graduate Women's Studies Conference, University of California at San Diego, La Jolla, CA
April 14-17, 1994 (April 16)

Cynthia A. Norman
UCSD Department of Communication

Not in Families Like Us: The Social Construction of Child Abuse in America

Most discussions of child abuse come from the realm of the professional: teachers, therapists, psychologists, health workers, etc. It is also a common topic in the popular press and the media. Within academia, however, there is surprisingly little work. Most of the discussions that do exist set up the problem of child abuse in ways that seem unproblematic. I would like to problematize the issues involved in child abuse and set up the problem in entirely different ways.

What I mean by the term child abuse here is violent acts done to a child on purpose by an adult (or sometimes another child). This encompasses physical violence, molestation, and emotional/verbal violence. To understand child abuse policy we must also add child neglect, a slippery category encompassing just about everything a parent can do that a person with political or social power over the parent contends is wrong (or the lack of parental action that said person contends is necessary). In the interests of time, most of my examples will be from one kind of abuse, sexual abuse.

In modern popular conceptions, child abuse is the violent act of deviant adults against powerless children Under this conception, the goal of society, therefore, is to ferret out such abuse whenever it is found and punish (or treat) the offenders. Because the perpetrators are deviant (from mental instability, criminal tendencies, lack of sexual control) they are routinely condemned and are (or should be) removed from the presence of their victims and potential victims.

This conception, however, is anything but innocently descriptive. It comes out of a series of shifting social tensions regarding the function of families, children, gender roles, sexuality, social class, race/ethnicity, and Americanism (among other things) in the structure of society at large. Child abuse began to be recognized as a social problem in the United States in the late nineteenth century, around the same time as the opening of the continent to travel and "instant" communication, the consolidation of extended families into nuclear families, vast changes in the structure of work under expanding industrialization, the Civil War and an end to legalized slavery, a larger concern with individual rights (including animal rights), and an enormous increase in immigration.

With the recognition of child abuse came a series of moves by the Anglo-American upper and middle classes to locate the idea of child abuse and other family violence in communities that were not their own. At the same time, Anglo volunteers, predominantly women, formed organizations to help women and children in the immigrant lower classes whose families were violent and/or alcoholic. This led to child protection agencies and later to the profession of social work.

Before going further with the history of child abuse intervention, I would like to make a few claims about child abuse itself. My largest claim is to say that child abuse (as well as other forms of abuse), as a system, is normal. Normal doesn't mean I accept the practice or am resigned to it. By normal I mean a set of practices embedded in daily life and inseparable from cultural norms required for the maintenance of certain structures of power. Abuse functions at many different levels, from reinscribing hierarchies of power between and among individuals to fortifying unequal economic, social, and political relationships between and among groups. If I say child abuse is normal, by this I mean that it works to maintain certain institutional structures that favor powerful groups. In order to see the normality of abuse, however, we must step away from a conception of it as an act one individual perpetrates upon another.

Among the systems that child abuse works within are binary oppositions such as gender, race, class, sexual orientation, etc. Binaries do not just describe difference, but hierarchy. The subordinate group is in a position of "otherness" with the dominant group. The dominant group derives its feelings of superiority from situating itself as not of the subordinate group. Men are not women, whites are not people of color, heterosexuals are not queer. How a person ends up in one group or another and where those dividing lines fit is sometimes based on tangible factors and sometimes is rather arbitrary. Also, the group that is "othered" is often composed of many sub-groups that make the hierarchical relationship more complex.

Binaries of ethnicity, class, and citizenship come into play when we look at the development of child protection agencies. The first American child protection workers were native-born middle and upper-class Anglo women in the 1870's and 1880's; their first clients were poor and working-class immigrant families. That these workers chose immigrants to help is no indication of where the problems of child abuse lay. While some immigrant families did abuse and neglect their children, so did some families of the native-born (including the families of the upper and middle classes). But poor parenting was firmly entrenched in the minds of many Anglos as a vice of the lower classes, along with alcoholism, violence, and other immoralities. Although the upper classes recognized these vices in themselves, and worked diligently to remove them, their presence did not create the same racial and class stereotypes as they did for families not of their class.

Far from being a "melting pot," America was an country forged from Western European stock, particularly British. All newcomers were expected to learn English and to accommodate themselves to an "American" culture. Naturally, some immigrants found this easier than others. But some Anglo-Americans felt that immigrants, and especially children of immigrants, could not only accommodate to American ways but transform themselves into full-fledged Americans. They needed only to be taught. Anglo women were at the forefront of the Americanization movement and they directed much of their attention to work with female immigrants.

Native-born work with immigrants (and the poor) took many forms, including poorhouses, public schools, and direct intervention into homes. And it sprang from many motivations, including "raising up" the workforce and creating an informed electorate. Because children were increasingly seen as a special class, more "savable" (and more innocent) than adults, charity directed at children grew in popularity. Public school programs were increasingly part of the efforts. David Herman, writing about California, argues:

The Americanization movement of the early twentieth century was originally intended as a benevolent expression of native-born, middle class concern for the plight of immigrant workers and their families.

Yet, as Herman and others make clear, Americanization satisfied many agendas. Foremost among them, I will argue, is the desire of the "ruling classes" to preserve their own privileges and prestige. They did this by identifying themselves as "Americans" (notably, not as Anglo-Americans), and therefore the norm for others to aspire to. Non-Americans were therefore defined as deviant. In this way, the upper classes could ignore in themselves the very behaviors that they sought to transform in others; including acts of family violence.

Creating an "other," people classified as "not-us," allowed the Anglo-Americans to project their own misdoings, on to a class clearly marked as outside themselves. Race and ethnicity were not the only axes on which to divide people. Women, children, servants and slaves also fulfilled roles as "others"; not simply for the mass of the (ruling) society, but as counter to the individual subject, husband, father, master. Children also play the role of "other" to their parents. Piagetian, and several other, theories of child development cast children's thinking as non-rational and concrete, closer to a natural state. But children and immigrants both transform themselves. They move from object to subject; their "othered" status is temporary. Children grow up to become adults. Their transformation is individual and expected. Immigrants sometimes make their transformation as individuals, more often they do it as a class, over generations. This transformation is achieved, not part of the "natural order" as is maturation. Not coincidentally, it is predominantly through the acculturation of children that immigrants' status changes.

Unfortunately, I don't have time to go into more detail about Americanization and its connection to family violence. Although such violence is no longer located in the groups of European immigrants, race and class still play major roles in how child protection agencies, as well as public perception of child abuse, operate. Upper and middle class Anglo-Americans, and European-Americans assimilated into Anglo culture, still escape the stereotype of family and other forms of violence now associated mainly with African-Americans, Hispanics, the poor and working classes, and those living outside the "safety" of the suburbs.

Binary oppositions are active with the act of child abuse as well. The prominent binaries here are gender, sexual orientation, and, perhaps most prominently, ageQthe binary of children vs. adults. The practice of child abuse as well as practices surrounding child abuse helps to re-inscribe relationships of power between these groups. Acts of violence are powerful practices that work to convey clear ideas of position or place. Feminists have argued that rape and domestic violence serve this purpose of maintaining relationships of power between men and women; scholars of lynching argue that this practice served to re-inscribe dominance of whites over blacks.

It would be a mistake, however, to look only to the surface of violent acts to find power hierarchies. Incest, for example, isn't just about children and adults but, usually, about fathers and daughters. Lynching served also to re-enforce standards of white Southern ladyhood, despite the fact that nearly all lynchings involved only black men and white men. Nor do we want to take forgranted the statistics. It is true that most rapes and incestual acts have male perpetrators and female victims. It is true that most adult domestic violence involves men beating women. But other configurations happen and are not uncommon. Particularly in the area of physical child abuse, male victims and female perpetrators are so common as to make the generalizations fall apart. Nor is all child abuse conducted by adults.

Instead of thinking of violence in terms of one group dominating another, let us look at violence as part of a network of relations. We must look beyond the actors to see the effects a violent act produces. In other work, I argue that child abuse, both sexual and physical, serves to reproduce gender norms, the institutions of masculinity and femininity, hierarchies of age, nuclear family structures, and the economic and social systems that intersect with nuclear families and patriarchy. These claims hold even when a specific act of abuse involves child or female perpetrators, male victims, or other non-obvious configurations. We can see violence as an act that not only produces effects (death, injury) but that sends messages, messages to the victim and to others as well. In this way, I see all forms of violence as discourse that assists in the construction of social systems.

Let us look in detail at some examples. Many forms of violence send very clear messages. Jacquelyn Dowd Hall, (in "The Mind that Burns in each Body: Women, Rape, and Racial Violence,") discusses lynching in the post-Reconstruction South and maintains that it was not necessary for lynching to be widespread for it to be effective; the fear of it was enough. Lynching was bound up with issues of rape as well since lynchings were justified as being revenge for the rape of a white woman by a black man (despite the fact that most lynching victims were never accused of rape). The message here was one of "place."

But the message was not intended for black men alone. According to Hall, lynching also communicated roles to women. White women were constructed as needing male protection from black "brutes." To give up being a lady opened white women up to attack; not only hypothetical ones by black men, but retaliatory ones by white men. White women were thought to be incapable of giving consent to sex with black men, black women were seen as being unrapable by any man; it could never be against her will.

The "victors'" use of rape in wartime to send a message of defeat to their opponents is another example of this. Rape also serves a local function of social control. Women know they should not walk alone at night because they might be attacked. They should not be unaccompanied in bars after a certain hour. They should not act too aggressively at work with male co-workers or subordinates. The messages in child abuse are somewhat different, but their means of communication remain similar.

Let's look more closely at the common view of child abuse and rape as acting within certain boundaries of classes, in this case, gender and age. Liberal feminists such as Florence Rush and Susan Brownmiller make these arguments about child sexual abuse and adult rape, respectively. Both argue for a kind of essentialism of gender and age categories. For child sexual abuse, children are likely victims, the argument goes, because they are physically small and trusting of (or at least indebted to) authority figures.

Rush, in her 1980 book, The Best Kept Secret: Sexual Abuse of Children, makes this argument in her study of child sexual abuse:

Inequality between a child and adult should alone forewarn of emotional problems, and the physical dangers inherent in difference in size and strength present clear-cut medical hazards. A tiny mouth, anus or vagina cannot accommodate an erect penis. (Rush, pp. 1-2).

Brownmiller has made the same argument about rape, that the physical difference between men and women alone (in both overall size and in penis vs. vagina) makes rape an inevitable danger. Not coincidentally, Brownmiller wrote the introduction to The Best Kept Secret and praises Rush for bringing about "the new nationwide interest in the sexually abused child." (Brownmiller, Introduction in Rush, pp. ix)

But these arguments assume something essential and obvious about gender categories. More importantly, they don't account for sexual abuse of boys (by men or women) or sexual or physical abuse by women (to girls or boys). Sexual abuse of boys by men is widely known, and Rush does discuss it. She maintains that when a boy is molested, he is "being used as a woman." Forced sex of a boy by a woman doesn't count in the same way because the boy is still (socially) male. She also quotes a woman who as a child was molested by a female teacher:

It really wasn't pleasant and I hated it but it was only one time. Being molested by men is scary because a man can kill you and besides it's something that goes on all your life. (Rush, p. 181)

Unfortunately, Rush takes at face value the assertions of molested boys or girls molested by women that the event had little effect (or less effect than if the molestation occurred between and man and a child). She never questions why descriptions of such acts might be construed in different ways because of larger issues of gender and power relations within the culture. The act is the same but the result is presumably different. Yet various books and other sources designed for adults molested as children, find that their litany of long-term emotional effects holds regardless of the gender of the perpetrator or victim.

Much of child sexual abuse is between siblings or other pairs where both victim and perpetrator are children. Of course, child-child sexual activity is often consensual exploration. Though, as Rush says, a large difference in age and/or physical size and strength can tip off an outsider that something is probably wrong (i.e., that the younger/smaller child is not consenting). But to focus purely on differences in size, strength, and age as removing the ability to consent, would be saying that most heterosexual (and many same-sex) adult relationships are inherently coercive. Since Rush is borrowing the argument from Brownmiller, it is not enough to say that Rush only applies it to children, that only children are inherently unable to consent to sex. Rush's application of the size and strength theory also falls apart when we see that she has assumed (as does Brownmiller) that all men have physical power over all women (by virtue of having a penis). This leads to her circular argument that childrenUs bodies are too small to withstand penetration by a penis so then all molestation is coercive, even when it does not involve penetration (or a penis).

An example of coercion that does not fit Rush's theory comes from an article in Child Abuse and Neglect on female child perpetrators of child molestation. In one case, a young girl forced, through emotional intimidation, her older and larger male cousin into engaging in sexual activity (including intercourse) with her. I argue that, instead of looking to physical difference to situate and explain sexual abuse, we should look at broader structures of power within the culture. Instead of matching up characteristics of individual actors engaged in sexual activity to see who is "on top," we should look at how the sexual act is used by the participants, and the larger society, to define and produce relationships of power. This same argument works for non-sexual types of abuse as well.

To say also that molestation (1) is a special category because it involves sex with children (an action that is all right between consenting adults) ignores acts of physical child abuse (where consent is never an issue). Again, these are both actions that adults (or occasionally children) do to children, usually within a family context. We can also see violence in relationships that are between adults such as the parent-child-like relationship servants and slaves were supposed to have with their employers/masters or between husbands and wives where there is a long history of expected inequality.

Child abuse is both common and taboo. In fact, it is too common to be dismissed as a product of individual madness or even as deviant (2). Like other social acts of violence it is "designed" to be devastating to the victim. It is not the behavior per se that is the issue but the meaning it carries. We can call this the act's "social intent" (in order to separate it from a sense of individual intent), although this phrase can be misleading. The act itself carries a certain meaning, one that is culturally constructed to be sure, but not necessarily a meaning that the perpetrator comprehends or even desires. The act itself works as a form of discourse that inscribes and reinscribes the power relationships involved in the transaction of the behavior.

An example of how this works can be shown in this excerpt of an interview with a female survivor of child sexual abuse who, as an teenager, molests a baby girl:

When I was fourteen I was baby sitting for a little girl who was around two. I was diapering her, and she was lying there with her legs spread, and I felt furious with her for how vulnerable she was. "You can't do that! You can't be a little girl in the world like that!" And what I ended up doing was touching her vagina and putting my finger inside. I did it for a minute or two, and I was furious at her the whole time. I hated her for being vulnerable. And I had this warped feeling that I was protecting her. "You can't go around being this vulnerable, so I'm going to do something to make you less vulnerable. That way, when the real child molesters come along, you won't be hurt by it."

I had the assumption this happened to every little girl, and that didn't stop me from hurting a child. I had to smash the vulnerability I saw in her. It frightened me. It was that simple (3).

Although the teenager's stated intent was (in a sense) one of protecting the child, she drew on an action which she knew would send a certain message to the child. The action itself did not do this; there is no indication that the child was physically hurt or even cried. Nor could the child have been completely aware at that age of what was happening to her. But just as the discourse she is a part of since birth constructs how she is female, a child, a daughter, her ethnicity, class, etc. and sends a message to that effect as to how she should behave in the world, the act of molestation also constructs a message, one of vulnerability and roles, of sexuality and relationships.

I do not believe we can separate sexual violence from physical violence, violence to adults from violence to children. Nor can we remove these violences from their contexts of race, gender, class, etc. Child abuse, though it does have some elements that set it off from other forms of violence, is, in other ways, very similar to other kinds of abuse. Child abuse, by definition, is a behavior directed at children, the youngest and smallest members of a family who need to learn their "place" in that structure. The message in an abusive family is clear: adults rule children (with the frequent corollary that husbands rule wives).

Child abuse, like any other social phenomenon, has both a history and a construction. Although the actions currently associated with child abuse have existed for a very long time, their association as a unit with social meaning did not happen in the United States until the 1870's when child abuse became both an entity reported in the media and one which triggered societal intervention.

What child abuse was and is, however, is completely inseparable from constructions of categorical difference, institutions, and power. This includes not only such directly related issues as constructions of childhood, parenting, and family prerogative but also issues of race, ethnicity, class, and gender. As child protection services solidified themselves into institutions, lines of class and ethnicity were carefully maintained. Even today, although the specifics of the groups associated with the violence/safety dichotomy often employed to justify large-scale intervention have changed, the distinct markings of class and ethnic struggle remain. Acts of child abuse help maintain the very fabric of a society woven by particular structures of family, gender, and "place."


  1. For purposes of clarity, I will use the term molestation to refer to abusive sexual acts performed on children (of either sex) and the term rape to refer to abusive/non-consensual sexual acts performed on adults (of either sex).
  2. Current estimates in the literature are that 40% of all women and 20-30% of all men in the United States were molested in some way at least once before the age of 18. Additional numbers of adults were physically abused or neglected as children (although many children suffer more than one kind of abuse/neglect).
  3. Bass and Davis, The Courage to Heal, pp. 285-286.

Cyndi Norman / / Last Modified: 12/20/95
 Return to Personal Main Page
 Return to Cyndi Norman's Homepage