1. reading RESEARCH ON THE EFFECTS OF CHILDHOOD ABUSE Perceptions of Juvenile Offenders Who Were Abused as Children From this article, found interesting or useful in some way. For this reading, prepar

KEYWORDS juvenile offenders, child abuse, expert testimony, child maltreatment, race, juvenile delinquency, juvenile sentencing, mitigating factors Child abuse has recently received ample attention from social scientists, although historically it has not always been recognized or addressed as a problem (Finkelhor, 1984). Within the past 15 to 20 years, researchers have Received 4 September 2007; revised 6 August 2008; revised 28 November 2008; accepted 16 March 2009.

I thank Dr. Bette Bottoms and Nancy Halbutzel for their helpful comments on the project.

Address correspondence to Margaret C. Stevenson, University of Evansville, Department of Psychology, 1800 Lincoln Avenue, Evansville, IN 47722. E-mail: [email protected] 332M. C. Stevenson studied a range of issues related to child abuse, including long-term psycho- logical effects of child abuse, causes of child abuse, and perceptions of child abuse. One well-documented finding in this field is a link between a history of child maltreatment and juvenile delinquency, such that many juvenile offenders have been victims of child abuse. It was Curtis (1963) who first articulated the now popular theory of the “cycle of violence” in a brief report where he worried that abused children might “become tomor- row’s murderers and perpetrators of other crimes of violence, if they survive” (p. 386). In fact, a recent issue of Children’s Services: Social Policy, Research, and Practice was devoted to this very topic (Quas, Bottoms, & Nunez, 2002). The link between child abuse and juvenile delinquency is so well-documented that Grisso (2002) expressed concern that clinicians testi- fying in court might incorrectly use the link between child abuse and juve- nile delinquency as evidence that abused juveniles will reoffend and should therefore be treated more punitively than their nonabused counterparts.

Grisso’s concern begs the question: Is child abuse used as a factor that makes the crime less severe, supporting a lenient sentence, or more severe, supporting a punitive sentence?

The Juvenile Court Act (1987) requires that a history of childhood phys- ical abuse be considered as a mitigating factor (i.e., facts that make the crime less serious, less intentional, and lead to lenient sentences) when determining whether to transfer juveniles from the juvenile justice system to the adult criminal court system, where they are tried by juries as adults. The adult criminal court system also mandates that juries be allowed to consider all relevant mitigating evidence, including nonstatutory evidence like an abusive family background, when determining the death penalty (Williams v. Taylor, 2000). Such laws play to natural attitudes toward justice and fairness: that an abused offender deserves more sympathy and, conse- quently, less severe treatment in comparison to a nonabused offender. In fact, some mock trial studies have shown that a juvenile defendant’s history of child abuse is used by jurors as a mitigating factor, eliciting less severe decisions (e.g., fewer transfers to adult court) as compared to comparable defendants with no history of abuse (Nunez, 2005; Nunez, Dahl, Tang, & Jensen, 2007; Stalans & Henry, 1994).

Are abused juvenile offenders actually treated less severely than non- abused juvenile offenders within the juvenile justice system? Some research suggests that this is not the case and that maltreated juvenile offenders, as compared to nonmaltreated juvenile offenders, are perceived by juvenile court officials as less likely to be rehabilitated and as more deserving of severe treatment (i.e., incarceration or transfer to adult court; Grisso, Tomkins, & Casey, 1988; Salekin, Yff, Neumann, Leistico, & Zalot, 2002). In those studies, however, child abuse is confounded with a host of other factors, including, for example, an unsupportive “chaotic” family environ- ment, juvenile behavioral problems (i.e., hostile demeanor), mental health Juvenile Offenders Abused as Children333 problems, and school problems, all of which have been found to predict puni- tive juvenile court sentences rendered by juvenile court officials (Clarke & Koch, 1980; Fenwick, 1982; Grisso et al., 1988; Horwitz & Wasserman, 1980).

The main goal of the current article is to present a review of all relevant research necessary to fully understand the complex relationship between childhood abuse and perceptions of juvenile offenders. Further, this review is intended to provide a unique understanding of some of the heretofore unexposed consequences of having a history of physical abuse within the justice system. That is, although mock jurors appear to use child abuse as a mitigating factor in controlled studies where it is experimentally manipu- lated, juvenile court officials do not seem to use it as a mitigating factor in real cases within the juvenile justice system. Instead, abused juvenile offenders appear to be treated more punitively than their nonabused coun- terparts. Why? Are there factors that covary with a history of child abuse in the real world that ultimately drive the seemingly aggravating effect of child abuse on juvenile sentences? This review is an attempt to resolve this issue by explaining these conflicting lines of research. Specifically, this review covers all empirical studies that address the following issues: (a) all pub- lished studies (seven total) examining how mock jurors use a defendant’s history of child abuse in determining their treatment of an alleged offender, (b) all existing studies (eight total) on juvenile court officials’ (i.e., probation officers and juvenile court judges) perceptions of actual juvenile offenders who have a history of abuse, and (c) a representative sampling of empirical research on factors that co-occur with abuse history, and research suggest- ing how these confounds might account for the inconsistencies in these two bodies of research. (These studies were identified by following basic litera- ture search procedures [i.e., using standard literature search databases like PsychINFO and Current Contents, searching for relevant research cited within relevant articles, contacting professionals who conduct research in these areas to identify any unpublished, yet relevant research]. These stud- ies were published between 1980 and 2008). PERCEPTIONS OF CHILD ABUSE AS A MITIGATING FACTOR To review research relevant to understanding potential jurors’ perceptions of child abuse as a mitigating factor, studies in which child abuse was not exper- imentally manipulated are first presented. This is followed by studies in which child abuse was experimentally manipulated in the context of a mock trial.

Studies in Which Child Abuse was not Experimentally Manipulated Heath, Stone, Darley, and Grannemann (2003) examined mock jurors’ ratings of the persuasiveness of 15 different “excuse defenses” (including a 334M. C. Stevenson history of child abuse) for a hypothetical crime in which an adult defendant was accused of assault and battery of an innocent victim. Participants also made case judgments and recommended a sentence for the defendant con- sidering each of the excuse defenses separately. They found three clusters of excuses that varied in level of perceived persuasiveness. The more persuasive the excuse, the less responsible participants perceived the defen- dant to be, the less control the defendant was believed to have over the crime, and the shorter the sentences were that the jurors recommended. The most persuasive cluster of excuses included a history of parent-perpetrated abuse and various genetic disorders, the second most persuasive cluster included different types of drug use and addiction, and the third cluster included excuses like violent television programming, subliminal messages in music, and nicotine withdrawal.

Lynch and Haney (2000) also provided some evidence that individuals perceive child abuse as a mitigating factor. They gave community members a death penalty case scenario that happened to include a history of abuse as a mitigating factor, as well as a number of other factors (abuse history was not experimentally varied). They then asked participants to indicate whether they perceived his child abuse as supporting a life sentence, a death sentence, or as not relevant. The majority of the participants (63%) indicated that they used the defendant’s history of child abuse as a mitigat- ing factor that supported a life sentence rather than a death sentence.

Approximately 25% of participants, however, indicated that a history of child abuse did not influence their decision, and approximately 13% stated that they actually used a history of child abuse as an aggravating factor, even though it was presented by the defense as a mitigating factor.

Garvey (1998) surveyed jurors who sat in actual capital murder cases about their beliefs regarding potential mitigating and aggravating factors. In contrast to Lynch and Haney’s (2000) findings, only approximately one-third of participants indicated that a defendant’s history of child abuse would mit- igate their judgments in a capital trial, while approximately two-thirds of the participants indicated that child abuse would not influence their judgments.

Yet child abuse was rated as a mitigating factor more than the majority of the other potentially mitigating factors in the survey, including, for example, no previous record, a background of poverty, alcoholism, or drug addiction.

In fact, jurors rated only 5 of 14 defendant characteristics as more mitigating than a history of child abuse, and those included the defendant being mentally retarded, having a mental illness, being younger than 18 years old, being institutionalized but not helped, and doubt of the defendant’s guilt.

One explanation for the low endorsement of child abuse as a mitigating fac- tor relative to Lynch and Haney’s study might be due to the fact that Lynch and Haney asked participants to make case judgments in reference to a spe- cific case transcript, whereas Garvey asked participants to complete a series of abstract survey questions. In fact, research has demonstrated that people Juvenile Offenders Abused as Children335 express a preference for more punitive treatment of offenders when they are asked abstract questions as compared to when they are asked questions specific to a given case (e.g., Stalans & Diamond, 1990).

In unpublished dissertation research, Stevenson et al. (2008) examined how jurors’ discuss a defendant’s history of child abuse while deliberating on a death sentence for an adult defendant. They presented jurors with a videotaped mock death penalty case and instructed them to deliberate in groups of 11 to 12 to reach a sentence (life or death). Testimony in the case included that the defendant had a history of childhood physical abuse. The jurors’ videotaped deliberations were transcribed and coded for their discus- sions of child abuse. Although jurors often used the defendant’s child abuse directly as a mitigating factor (e.g., “His child abuse explains his behavior”), the full story was more complex than this. That is, in addition to directly dis- cussing child abuse as a mitigator, jurors even more frequently argued to discount the abuse as a mitigating factor (e.g., “Plenty of kids have been abused and they don’t go on to commit murder”) and sometimes even directly used it against him as an aggravating factor (e.g., “I think the abuse is more of a predictor—when someone’s been treated that way, they’re more likely to be that way”). In fact, jurors were significantly more likely to discount his abuse as a mitigator or even directly use it as an aggravator than they were to use it as a mitigating factor (an argument for a life sen- tence). Some jurors even went as far as to argue the following: “Maybe the damn kid deserved it. Maybe that’s why his dad beat the hell out of him.” These results might surprise those who intuitively assume that jurors would use a defendant’s history of child abuse to argue for a life sentence.

In summary, although these studies did not directly assess the extent to which child abuse influenced jurors’ case judgments (via experimental manipulation), they did reveal interesting insight into jurors’ use of child abuse. That is, although mock jurors appear to think that they would use child abuse as a mitigating factor, Stevenson and colleagues (2008) show that they more frequently do not.

Vignette Studies in Which Child Abuse was Experimentally Manipulated Studies in which a juvenile offender’s history of child abuse was experimen- tally manipulated within the context of a case vignette provide another important perspective regarding jurors’ use of child abuse (Nunez, 2005; Nunez et al., 2007; Stalans & Henry, 1994). In all three studies, the authors predicted that a history of child abuse would elicit less severe case judg- ments relative to no history of child abuse. Stalans and Henry (1994) relied on attribution theory as a basis for their hypotheses. In general, theories of attribution suggest that participants will render more punitive case judg- ments when the cause of the crime is perceived to have been internal 336M. C. Stevenson (i.e., caused by factors within the person, like disposition), controllable (i.e., intentional), and stable (i.e., caused by a constant factor and is therefore likely to occur again), as opposed to external (i.e., caused by situational factors, like the environment), uncontrollable (i.e., unintentional), and unstable (i.e., the event is unlikely to occur again; Shaver, 1985; Weiner, 2006). In fact, research has shown that the lower the perceived internality, intentionality, and stability of the cause of the crime, the less punitively par- ticipants treat an offender (Carroll, 1978; Carroll & Payne, 1977; Graham, Weiner, & Zucker, 1997). Thus, to the extent that child abuse is perceived as an external cause of the crime and that an abused juvenile offender is perceived as having less intent to commit a crime and less understanding that crime is wrong than a nonabused juvenile, participants should render less punitive case judgments for an abused versus nonabused juvenile offender.

As a test of this hypothesis, Stalans and Henry (1994) explored the influence of a history of child abuse on mock jurors’ perceptions of a juve- nile and recommendations that the juvenile be transferred to adult court (i.e., the most severe outcome). They recruited community members from Georgia to read a series of short vignettes describing a 16-year-old boy as having killed either his father or neighbor after a heated argument. The two factors that were varied between subjects were whether or not the boy had been repeatedly physically abused (by the father) and whether the victim was a neighbor or the father. When the boy had not been a victim of child abuse, compared to when he had been a victim, participants were signifi- cantly more likely to believe that the boy understood the wrongfulness of his actions, that he intended to kill the victim, that transferring the boy to adult court would deter others from crime, and that the boy should be transferred to adult criminal court. But the effect of child abuse on recom- mendations for adult court was qualified by an interaction indicating that participants were least likely to recommend a juvenile who killed an abu- sive father to adult court than any other condition, including an abused juvenile who killed a neighbor. Thus it is possible that a “self-defense” explanation explains, in part, why participants were most lenient toward an abused juvenile who killed his father. That is, jurors likely believed the abused juvenile felt more threatened and reacted in self-defense when the victim was his previously abusive father than his nonabusive neighbor.

Even so, participants were more likely to recommend adult court when a nonabused juvenile killed a neighbor than when an abused juvenile killed a neighbor, demonstrating that child abuse ultimately was still used as a miti- gating factor even in the absence of a self-defense explanation (i.e., when the victim is the juvenile’s nonabusive neighbor).

Nunez et al. (2007) conducted a similar study exploring the influence of child abuse and juvenile offender age (11, 13, or 15 years old) on jurors’ recommendations to transfer a female or male juvenile to adult court. They Juvenile Offenders Abused as Children337 found that a history of child physical abuse (compared to no history of child abuse) predicted fewer recommendations for transfers to adult court for both male and female juvenile offenders accused of murder. In other words, a history of child abuse was used as a mitigating factor for both male and female juvenile offenders. Fifteen-year-old boys with no history of child abuse, however, were far less likely than their female counterparts to be recommended to juvenile court. Nunez et al. also varied the victim of the juvenile offender from the father to the neighbor and found some effects of victim type similar to those found by Stalans and Henry (1994). For exam- ple, a juvenile who killed a father was twice as likely as a juvenile who killed a neighbor to be recommended for juvenile court. Still, the mitigating effect of child abuse on transfers to adult court was significant both when the neighbor was the victim and when the father was the victim. These find- ings offer further support that a history of child abuse is still used as a miti- gating factor even in the absence of a “self-defense” argument.

Finally, using a different vignette of a juvenile accused of murder, Nunez (2005) varied a history of child abuse and the gender of the juvenile offender, but the victim was always the juvenile’s father. Again, Nunez found that a history of child abuse was used as a mitigating factor such that nonabused juveniles received more severe verdicts (i.e., first-degree murder) than abused juveniles. Because this study included only the abu- sive father as the victim (and not a neighbor), the implications are limited to a “self-defense” argument, yet child abuse was used as a mitigating factor nonetheless. Interestingly, in contrast to Nunez et al.’s (2007) findings, Nunez (2005) found that girls accused of murdering a father were given more severe verdicts than boys. This main effect of gender, however, was qualified by a two-way interaction of gender and abuse history such that a history of abuse was used as a mitigating factor more for boys than for girls.

One potential explanation might be that such violent crimes are associated with boys more than girls due to media exposure and gender stereotypes dictating that women are less aggressive and more passive and submissive than men (Deaux & Lewis, 1984). Thus a girl who commits a violent crime might be perceived as particularly aggressive and deviant because she is violating gender norms, as opposed to a boy, which, in turn, might lead jurors to be less influenced by a mitigating factor of child abuse.

Why, though, did Nunez et al. (2007) find few differences in the treatment of male and female juvenile offenders, whereas Nunez (2005) found that girls were treated more severely than boys? Whereas both case scenarios depicted a violent murder (i.e., the juvenile shot the victim to death), Nunez (2005) employed a much longer and more detailed crime scenario (i.e., the juvenile planned the murder with detail, planted many weapons around the father’s house, and waited patiently for him to come home to shoot him to death), as opposed to a short paragraph used in the study of Nunez et al. (2007). Thus the greater realism in Nunez’s (2007) 338M. C. Stevenson study might be the reason the same effects were not found by Nunez et al.

(2007).

Perhaps gender stereotypes that depict women as less aggressive and more passive and submissive than men (Deaux & Lewis, 1984) actually cause women to be perceived as particularly deviant and aggressive when they commit violent crimes. To the extent that gender acts as a proxy variable for juvenile offender demeanor (with girl offenders perceived as more hostile than boy offenders), gender might influence the mitigating effects of child maltreatment on juvenile sentences. That is, when the juvenile offender is female, as opposed to male, a history of child abuse might be less likely to be used as a mitigating factor and more likely to be used as an aggravating factor. Thus juvenile gender is an interesting and important factor to consider when interpreting the influence of a history of child abuse on case decisions. JUVENILE COURT OFFICIALS’ PERCEPTIONS OF ABUSED JUVENILE OFFENDERS The vignette studies reviewed thus far have shown that mock jurors, on average, use a juvenile offender’s history of child abuse as a mitigating factor when it is experimentally manipulated. Research reviewed next, how- ever, reveals that juvenile court officials do not use child abuse as a mitigat- ing factor, and instead use it as an aggravating factor. Why would juvenile court officials behave differently than mock jurors? One might argue that there are differences in the ways in which laypersons versus juvenile court officials use a history of child abuse when rendering juvenile dispositions.

Perhaps there is something unique about juvenile court officials’ experience with juvenile offenders that might cause them to not use child abuse as a mitigating factor. For instance, perhaps court officials’ schemas (i.e., mental representations) of juveniles who are likely to recidivate include juveniles with a history of child abuse. Thus juvenile court officials might be more likely than novices to use a history of child abuse as a cue that influences their perceptions of the juvenile offender, perhaps even facilitating informa- tional inferences or assumptions about the juvenile based on information not directly provided. Such inferences might consist of charactaristics that correspond with perceived risk for recidivism and need for incarceration (e.g., negative demeanor), leading, in turn, to punitive sentencing.

However, as reviewed in this article, an alternative and more parsimoni- ous explanation for these conflicting lines of research is more likely: These two lines of research differ in methodology, topic, and goals. Whereas the vignette studies were aimed at understanding mock jurors’ perceptions of juvenile offenders, studies reviewed next were aimed at understanding juve- nile court officials’ perceptions of juvenile offenders. Thus, although it is cer- tainly informative to understand how a history of child abuse influences mock Juvenile Offenders Abused as Children339 jurors’ decisions, it would be misleading to rely solely on this line of research when attempting to understand how a history of child abuse is used within the juvenile justice system. In reality, juvenile justice decision making is com- plex and is influenced by a number of factors. Thus, as opposed to experi- mentally controlled studies, studies conducted in a natural setting can take into account the complex ways in which child abuse might influence juvenile court officials’ perceptions of juvenile offenders. Yet with more realism, of course, comes greater limitations in terms of cause-and-effect conclusions.

That is, to study the effects of child abuse on jurors’ decisions in a realistic environment, it is necessary to include factors that naturally covary with child abuse, which, in turn, act as experimental confounds that severely limit inter- nal validity (i.e., ability to determine causation).

Surveys of juvenile court officials conducted by Grisso et al. (1988) and Sanborn (1996) suggest that, surprisingly, child abuse appears to be used by juvenile court officials as an aggravating factor, leading to more severe treat- ment. Why might child abuse be used as a mitigating factor by laypersons within the context of controlled vignette studies, but not by actual juvenile court officials within the juvenile justice system? Several factors that are linked to child abuse in the real world (natural confounds of child abuse) were not presented in controlled experiments (e.g., a problematic demeanor, family environment, mental health, and inadequate school progress). Real juvenile court officials face abused children who tend to have many of these negative characteristics, such as behavioral problems, school problems, and an unsupportive family environment that does not foster rehabilitation. Mock jurors, however, never saw those characteristics in the experimentally controlled studies, and rightly so—they would repre- sent methodological confounds, making it difficult to determine whether the resulting effects were due to the childhood abuse or to those additional characteristics. Next, I review these characteristics that covary with child abuse and discuss how they might be responsible for the apparent aggravat- ing effects of child abuse within the juvenile justice system.

Confounds of Child Abuse One confound of child abuse not accounted for by the previously reviewed vignette studies is juvenile demeanor. That is, abused children are more likely than nonabused children to have anger management problems, antisocial behavioral problems, and poor social and emotional skills (Azar, Ferraro, & Breton, 1998; Cicchetti & Lynch, 1995; Egeland, Yates, Appleyard, & van Dulmen, 2002; Malinosky-Rummell & Hansen, 1993). In fact, the rela- tionship between child abuse and juvenile behavioral problems and crime has received significant empirical attention (e.g., Egeland et al., 2002; for a review, see Widom & Wilson, in press). Likewise, child abuse is also confounded with poor school performance. That is, studies have shown that 340M. C. Stevenson child abuse predicts delayed language and cognitive development and, in turn, sometimes results in subsequent poor school performance (for a review, see Veltman & Browne, 2001). In addition, research has shown that child abuse is a predictor of mental illness, such as depression and psycho- pathology (Bemporad & Romano, 1992; Luntz & Widom, 1994).

Yet another factor that is confounded with child abuse in real life is a chaotic family environment. Abused children are more likely than nonabused children to have parents who (a) lack child care knowledge, (b) lack parenting skills, (c) lack formal education, (d) face life stressors, (e) are at risk for depression, (f) are economically disadvantaged, (g) abuse drugs, and (h) live in dangerous and poor home environments (e.g., Kotch et al., 1997; Scannapieco & Connell-Carrick, 2005). It is important to note, however, that although there is a relationship between child abuse and these characteris- tics, this does not imply that every abused child will have every one (or even any) of these problems. Instead, these characteristics are simply more likely to be found in abused than nonabused children.

Surveys and Correlational Studies of Juvenile Court Officials Reviewed next is research indicating that juveniles with hostile demeanors, chaotic families, school problems, and mental health problems (factors that accompany child abuse in real life) are treated more severely by court offi- cials than juveniles without those problems, even when controlling for crime severity (Fenwick, 1982; Grisso et al., 1988; Horwitz & Wasserman, 1980; Leiber & Mack, 2003; Salekin et al., 2002; Sanborn, 1996). Because these negative factors that predict severe sentences covary with child abuse, this research will help explain why child abuse does not appear to mitigate case decisions in the juvenile justice system as it does in controlled studies, where these negative factors are not present. For the purposes of this arti- cle, a juvenile’s hostile demeanor, chaotic family, school problems, and mental health problems are conceptualized in terms of juvenile court offi- cials’ ratings and perceptions rather than clinical conceptualizations. Factors that influence juvenile court officials’ perceptions of juvenile offenders and sentences for juvenile offenders are typically studied in two different ways:

(a) surveys of juvenile court officials who rate various factors that influence the severity of juvenile sentences that they render, and (b) correlational studies that demonstrate the relation between actual juvenile sentences and various factors related to juveniles’ backgrounds.

S URVEYS OF JUVENILE COURT OFFICIALS Grisso et al. (1988) surveyed 1,423 juvenile court officials, including judges, prosecuting and defense attorneys, intake workers, probation officers, and mental health officials. When asked to indicate characteristics describing a Juvenile Offenders Abused as Children341 juvenile that should be incarcerated because he or she could not be rehabili- tated, court officials’ indicated the following factors: (a) low acceptance of intervention, (b) poor behavioral compliance, (c) past history of offenses, (d) poor academic functioning, (e) an unsupportive family, (f) negative peer influence, (g) an unsocialized family, (h) the presence of a mental disorder, and (i) high self-reliance. Note that all of these factors that court officials asso- ciate with a need for incarceration are also factors that covary with child mal- treatment. Thus it is likely the case that maltreated children are treated more severely than nonmaltreated children because of a host of additional prob- lems in those children’s lives that court officials believe (rightly or wrongly) lead to recidivism and should be headed off via incarceration. Also note that the factor labeled as “unsocialized family” indicates that the juvenile had a history of child abuse or neglect and a family environment of “chaos and dis- organization.” Thus this survey shows that court officials rated a history of child maltreatment itself as an indicator that the juvenile is unable to be reha- bilitated and deserves incarceration. Although court officials did specify child maltreatment as an indicator of a lack of rehabilitation potential, it is still pos- sible that factors that covary with child maltreatment (e.g., a chaotic family) drive this association between child abuse and punishment. That is, court offi- cials’ experience with juvenile offenders might have caused them to associate a history of child abuse with other negative factors and, in turn, begin per- ceiving child maltreatment itself as a predictor of recidivism. In other words, although child maltreatment is associated with incarceration, it is possible that the association is mediated by underlying factors that covary with child mal- treatment (i.e., a chaotic and unsupportive family, behavioral problems, etc.).

Sanborn (1996) found that the factor most frequently cited by juvenile court officials as being relevant to their sentencing decision was “the family,” specifically, the parents’ capability and willingness to control, supervise, and rehabilitate their child. Remaining factors included the par- ent’s character and the juvenile’s school performance, character, and mental health. The “parent’s character” was defined as evidence of physical abuse or neglect of their child, as well as their receptivity toward court interven- tions for their child. This again indicates that child abuse (although confounded with a perceived chaotic family environment) is used as a factor that court officials believe should influence sentence severity. The “juvenile’s character” was defined as the juvenile’s motivation to improve his or her behavior, the juvenile’s remorse for the crime, and his or her attitudes toward the court and authority figures. When asked the degree to which these factors are actually used when reaching sentences, 76% of the court officials agreed that most or all of these factors were used (Sanborn, 1996).

Salekin and colleagues (2002) asked judges to recall a case in which they decided to transfer a juvenile to adult criminal court and then judge the importance of a variety of the juvenile’s characteristics for determining the transfer decision. Consistent with the research just reviewed, the juvenile’s 342M. C. Stevenson demeanor (i.e., hostile personality, remorselessness, lack of empathy for the victim) and family environment (i.e., stable/supportive family) again emerged as factors predicting whether a juvenile is perceived as amenable to treatment and whether he or she should be transferred to adult court.

C ORRELATIONAL STUDIES Research on factors that predict actual juvenile sentences suggests that the officials’ in the studies just reviewed were accurate in their perceptions of what influences their decisions. Horwitz and Wasserman (1980) found that juvenile offenders with familial or school problems, regardless of the sever- ity of the crime, received more severe sentences (i.e., institutionalization) than juveniles without familial or school problems. To illustrate, Horwitz and Wasserman found that, of the juveniles who were institutionalized, practically all of them (92%) had a history of family or school problems.

In a study conducted by Fenwick (1982), observers systematically coded 350 juvenile intake hearings and examined the juveniles’ official court docu- ments. They measured the juveniles’ degree of “family disaffiliation” by indicat- ing factors such as the nature of the juvenile’s relationship with his or her family and whether both parents raised the juvenile. Even when controlling for crime seriousness, family disaffiliation predicted petitioning the youth to court (the most severe outcome following an intake hearing). Also, youths rated as having a negative demeanor (i.e., lacked respect, disinterested, and hostile) were more likely to be petitioned to court than youths with a positive or neu- tral demeanor, even while controlling for crime severity. Clarke and Koch (1980) also found that harsher sentences were given to juveniles whose par- ents did not attend the court hearing than to juveniles whose parents attended the court hearing, even after controlling for offense severity and prior offenses.

Vignette Study in Which Child Abuse was Experimentally Manipulated Vidal and Skeem (2007) conducted the only study thus far in which a his- tory of child abuse was experimentally manipulated and in which juvenile court probation officers served as participants. Officers perceived an abused juvenile as significantly more dangerous than a nonabused juvenile. They were also more likely to recommend secure residential placement, expect difficulty in supervision, and “go the extra mile” in their approach toward monitoring the juvenile by, for example, recommending psychological ser- vices more to an abused than a nonabused juvenile. Thus Vidal and Skeem (2007) present a more nuanced picture of the juvenile court officials’ per- ceptions of abused juvenile offenders. They expect that care for abused juveniles will be more difficult than nonabused juveniles and recommend Juvenile Offenders Abused as Children343 more serious institutionalization as a result, yet their sympathy for abused juveniles manifests in their greater likelihood to recommend psychological services to abused juveniles. SUMMARY Research reviewed thus far has shown that in vignette studies of laypersons’ perceptions of juveniles, child abuse elicits lenient case judgments (Nunez, 2005; Nunez et al., 2007; Stalans & Henry, 1994), but in surveys and correla- tional studies exploring juvenile court officials’ perceptions and case judg- ments, child abuse appears to yield punitive case judgments (Fenwick, 1982; Grisso et al., 1988; Horwitz & Wasserman, 1980; Salekin et al., 2002; Sanborn, 1996), more secure placement of the juvenile, and anticipated difficulties in rehabilitation (Vidal & Skeem, 2007). It appears that this discrepancy is explained by the fact that the real juveniles seen by officials exhibit a host of characteristics confounded with abuse in real life: chaotic family, hostile demeanor, mental health problems, and school problems— all of which mediate the link between abuse and severe sentences.

Although court officials might feel sympathy toward abused juveniles, a history of child abuse is not always used as a mitigating factor (as laboratory vignette studies suggest), but rather, a history of child abuse might indirectly be used as an aggravating factor within the juvenile court, perhaps in part due to the negative factors that covary with child abuse. It is likely that jurors would act similarly if faced with the same circumstances as juvenile court officials. In fact, Stevenson et al. (2008) found that when jurors were faced with an adult defendant who not only had a history of child abuse, but was also raised in a chaotic family environment with little parental support, they did not use his history of child abuse in a fully mitigating way. They were actually more likely to discount it or even use it against him as an indica- tor that he would commit future crime than they were to use it as a mitigator. FUTURE RESEARCH Thus child abuse is used as a mitigating factor only when manipulated separately from natural confounds of juvenile demeanor, family environ- ment, school problems, and mental health problems. Although I have pro- posed that confounds like a chaotic family environment ultimately explain why child abuse is not used as a mitigating factor in juvenile courts, this assumption has yet to be empirically tested and is a proposition that deserves future empirical attention.

For example, if abused juvenile offenders are treated more severely in juvenile court due to the fact that abused juveniles are more likely to have a 344M. C. Stevenson chaotic family environment, one could compare the sentences of juvenile offenders who are expected to return home with the originally abusive par- ents to the sentences of those who are expected to live with new, nonabu- sive parents, while controlling for crime severity. One would expect that juvenile court officials would be more likely to incarcerate a juvenile who currently lives with his or her abusive parents than a juvenile who lives with new, nonabusive parents. Why? The juvenile who lives with nonabusive parents might be perceived as having greater potential for rehabilitation than the juvenile who lives with the abusive parents. This hypothesis could be tested by obtaining actual juvenile court records, or more cleanly tested by experimentally manipulating the juvenile offender’s current caretakers within the context of a mock trial of a juvenile crime. Of course, the unique effects of other confounds of child abuse (e.g., behavioral problems, school problems, mental illnesses) could also be teased apart and explored via experimental manipulation.

Additional Moderators It is a fruitful exercise to consider other factors that might change the relation- ship between child abuse and juvenile sentence severity (i.e., moderators).

Although there are likely many, one is of particular societal importance:

juvenile offender race. Unlike Whites, Black juvenile offenders are overrep- resented in the juvenile justice system relative to their total population (Snyder, 2004). Even after controlling for offense severity and prior offenses, Black juveniles are more likely than White juveniles to be detained, trans- ferred to criminal court, and given longer sentences (Engen, Steen, & Bridges, 2002; Wordes, Bynun, & Corley, 1994). The simplest theoretical explanation for these findings is that of modern or aversive racism (Gaertner & Dovidio, 1986; McConahay, 1986; Sears, 1988). According to the theory of aversive racism, it is no longer socially acceptable to appear outwardly rac- ist (Gaertner & Dovidio, 1986). Although aversive racists endorse egalitarian values (which have become societal norms), their behavior is still influ- enced by anti-Black sentiments, as evidenced by research demonstrating that Black adult offenders, on average, receive longer sentences than White offenders in mock trial simulations (Sweeney & Haney, 1992). Thus people might be less likely to use child abuse as a mitigating factor for Black than for White juvenile offenders because anti-Black stereotypes (Devine, 1989) and attitudes (e.g., Gaertner & Dovidio, 1986) might elicit less sympathy and empathy for Black than for White juveniles. In support, Lynch and Haney (2000) found that jurors were not only less likely to use a history of childhood abuse as mitigating evidence for a Black than for a White adult defendant, but jurors were also more likely to incorrectly use some mitigating evidence (e.g., the defendant was under the influence of alcohol) as aggravating evidence when the defendant was Black than when the defendant was White. These Juvenile Offenders Abused as Children345 findings are supported by empirical evidence of anti-Black bias in case- related judgments of juvenile and adult offenders (e.g., Engen et al., 2002; Stevenson & Bottoms, in press; Sweeney & Haney, 1992; Wordes et al., 1994).

Thus future research should explore juvenile defendant race as a potential moderator of the effect of child abuse on juvenile sentence sever- ity. Additional potential moderators should be considered as well, including juvenile court officials’ attitudes toward juvenile offender rehabilitation potential and attitudes toward juvenile crime in general. Juvenile defendant gender also deserves future research, particularly because there currently is a surprisingly limited amount of empirical research regarding this factor, and what does exist reveals mixed findings, as was previously reviewed (Nunez, 2005; Nunez et al., 2007). PRACTICAL IMPLICATIONS The present review not only illuminates an unexpected consequence of a history of child abuse in the juvenile justice system, but it also has the potential to implicate useful applications designed to improve the legal system. For example, understanding how factors that co-occur with child abuse might be used against a juvenile provides information relevant for attorneys when they decide how to prepare and present their cases to jurors or juvenile court officials. Because the Juvenile Court Act (1987) specifies that a history of abuse be used as a mitigating factor in determin- ing whether to transfer a juvenile to adult court, attorneys representing juveniles must be aware that court officials do not intuitively use this information as mitigating evidence. In fact, studies reviewed herein have shown that abused juveniles are more likely to be recommended to adult court than nonabused juveniles (Grisso et al., 1988; Sanborn, 1996). Thus, to better serve justice and in keeping with the Juvenile Court Act, it might behoove attorneys to consider employing an expert psychological witness to testify about the psychological consequences of child abuse when determining whether an abused juvenile should be transferred to adult court. Informing decision makers of the many consequences of child abuse might lessen undue punitive treatment of abused juvenile offenders and cause them to be more likely to use a history of abuse as the law has prescribed: as a mitigating factor.

In fact, previous research has shown that attorneys and expert wit- nesses do influence jurors’ perceptions of the defendant and their verdicts (Diamond & Casper, 1992; Haegerich & Bottoms, 2000). Furthermore, it is likely the case that the way in which child abuse is presented or framed by an expert psychological witness will influence whether child abuse is used as a mitigating or an aggravating factor. As Grisso (2002) worried, expert psychologists might use the relationship between child maltreatment and 346M. C. Stevenson juvenile delinquency as evidence in a juvenile court hearing that child maltreatment predicts juvenile offender recidivism. Such an argument might cause the court to use child abuse as aggravating, rather than mitigating evidence. To the extent that expert psychologists persuade decision makers to believe a juvenile who has been physically abused as a child is more likely to be a chronic offender than a nonabused child, decision makers might perceive the juvenile as likely to recidivate and more of a threat to society, and, in turn, render a more severe sentence. In contrast, if an expert witness testifies that a juvenile offender who has been abused is particularly vulnerable and deserving of extra care and sympathy, decision makers might, in turn, render a more lenient sentence. Future research should test the possibility that an expert’s testimony on child abuse might be framed in such a way as to cause jurors to use it as aggravating rather than as mitigat- ing evidence.

Although child abuse appears to be used as an aggravating factor, elic- iting more punitive treatment in the juvenile justice system, it appears that court officials have sympathy for abused juvenile offenders and make con- certed efforts to aid their rehabilitation (Vidal & Skeem, 2007). Perhaps abused juvenile offenders might be best served by facilitating even more nonpunitive venues for their rehabilitation via psychological services. This recommendation is particularly warranted in light of research demonstrating that treatment programs emphasizing strict monitoring and control over juveniles actually adversely affect the rehabilitation of offenders and does not reduce recidivism (Paparozzi & Gendreau, 2005). Thus future lawmak- ers should consider incorporating specific goals that emphasize psychologi- cal treatment aiding the rehabilitation of abused juvenile offenders to help break the negative cycle that child abuse appears to perpetuate: maladap- tive behavior, delinquency, and, in turn, punitive treatment in juvenile court. CONCLUSION In this review I have questioned the assumption made by the courts and by researchers that a history of child abuse is used as a mitigating factor for juvenile offenders in the juvenile justice system. It is sobering that child abuse appears not to be used as a mitigating factor because it is con- founded with other factors that tend to yield more severe sentences. The unfortunate reality appears to be that those who are victimized as children face even greater disadvantages within the juvenile justice system relative to their nonvictimized counterparts. Future research on this topic is necessary to help ensure that those who are victimized as children are not being fur- ther victimized within the juvenile justice system. Fortunately this review provides fertile ground for specific empirical tests that not only address this Juvenile Offenders Abused as Children347 vital issue, but that will also advance the field of research in justice decision making. REFERENCES Azar, S. T., Ferraro, M. H., & Breton, S. J. (1998). Intrafamilial child maltreatment. In T. H. Ollendick & M. Hersen (Eds.), Handbook of child psychopathology (pp. 483–504). New York: Plenum Press.

Bemporad, J. R., & Romano, S. J. (1992). Childhood maltreatment and adult depres- sion: A review of research. In D. Cicchetti & S. L. Toth (Eds.), Rochester Sympo- sium on Developmental Psychopathology: Vol. 4. Developmental perspectives on depression (pp. 351–376). Rochester, NY: University of Rochester Press.

Carroll, J. S. (1978). Causal attributions in expert parole decisions. Journal of Personality and Social Psychology, 36, 1501–1511.

Carroll, J. S., & Payne, J. W. (1977). Crime seriousness, recidivism risk, and causal attributions in judgments of prison term by students and experts. Journal of Applied Psychology, 62, 595–602.

Cicchetti, D., & Lynch, M. (1995). Failures in the expectable environment and their impact on individual development: The case of child maltreatment. In D. J.

Cohen (Ed.), Developmental psychopathology: Risk, disorder, and adaptation (pp. 32–71). Oxford, England: John Wiley & Sons.

Clarke, S. H., & Koch, G. G. (1980). Juvenile court: Therapy or crime control, and do lawyers make a difference? Law & Society Review, 14, 263–308.

Curtis, G. C. (1963). Violence breeds violence—perhaps? American Journal of Psychiatry, 120, 386–387.

Deaux, K., & Lewis, L. L. (1984). The structure of gender stereotypes: Interrelation- ships among components and gender label. Journal of Personality and Social Psychology, 46, 991–1004.

Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56, 5–18.

Diamond, S. S., & Casper, J. D. (1992). Blindfolding the jury to verdict consequences:

Damages, experts, and the civil jury. Law & Society Review, 26, 513–563.

Egeland, B., Yates, T., Appleyard, K., & van Dulmen, M. (2002). The long-term con- sequences of maltreatment in the early years: A developmental pathway model to antisocial behavior. Children’s Services: Social Policy, Research, and Practice, 5, 249–260.

Engen, R. L., Steen, S., & Bridges, G. S. (2002). Racial disparities in the punishment of youth: A theoretical and empirical assessment of the literature. Social Problems, 49, 194–220.

Fenwick, C. R. (1982). Juvenile court intake decision making: The importance of family affiliation. Journal of Criminal Justice, 10, 443–453.

Finkelhor, D. (1984). Child sexual abuse: New theory and research. New York: Free Press.

Garvey, S. P. (1998). Aggravation and mitigation in capital cases: What do jurors think? Columbia Law Review, 98, 1538–1576. 348M. C. Stevenson Gaertner, S. L., & Dovidio, J. F. (1986). The aversive form of racism. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination and racism (pp. 61–90).

Orlando, FL: Academic Press.

Graham, S., Weiner, B., & Zucker, G. S. (1997). An attributional analysis of punish- ment goals and public reactions to O. J. Simpson. Personality and Social Psychology Bulletin, 23, 331–346.

Grisso, T. (2002). Using what we know about child maltreatment and delinquency.

Children’s Services: Social Policy, Research, and Practice, 5, 299–305.

Grisso, T., Tomkins, A., & Casey, P. (1988). Psychosocial concepts in juvenile law.

Law and Human Behavior, 12, 403–437.

Haegerich, T., M., & Bottoms, B. L. (2000). Empathy and jurors’ decisions in patri- cide trials involving child sexual assault allegations. Law and Human Behavior, 24, 421–448.

Heath, W. P., Stone, J., Darley, J. M., & Grannemann, B. D. (2003). Yes, I did it, but don’t blame me: Perceptions of excuse defenses. Journal of Psychiatry & Law, 31, 187–226.

Horwitz, A., & Wasserman, M. (1980). Formal rationality, substantive justice, and dis- crimination: A study of a juvenile court. Law and Human Behavior, 4, 103–115.

Juvenile Court Act. 705 ILCS 405/5-805, 5-810, 5-130 (1987).

Kotch, J. B., Browne, D. C., Ringwalt, C. L., Dufort, V., Ruina, E., Stewart, P. W., & Jung, J. (1997). Stress, social support, and substantiated maltreatment in the second and third years of life. Child Abuse & Neglect, 21, 1025–1037.

Leiber, M. J., & Mack, K. Y. (2003). The individual and joint effects of race, gender, and family status on juvenile justice decision-making. Journal of Research in Crime and Delinquency, 40, 34–70.

Lynch, M., & Haney, C. (2000). Discrimination and instructional comprehension:

Guided discretion, racial bias, and the death penalty. Law and Human Behavior, 24, 337–358.

Luntz, B. K., & Widom, C. S. (1994). Antisocial personality disorder in abused and neglected children grown up. American Journal of Psychiatry, 151, 670–674.

Malinosky-Rummell, R., & Hansen, D. (1993). Long-term consequences of child- hood physical abuse. Psychological Bulletin, 114, 68–79.

McConahay, J. B. (1986). Modern racism and modern discrimination: The effects of race, racial attitudes, and context on simulated hiring decisions. Personality and Social Psychology Bulletin, 9, 551–558.

Nunez, N. (2005, March). Juror perceptions of juveniles who commit murder: Are female defendants at a disadvantage? Paper presented at the annual meeting of the American Psychology-Law Society, La Jolla, CA.

Nunez, N., Dahl, M. J., Tang, C. M., & Jensen, B. L. (2007). Trial venue decisions in juvenile cases: Mitigating and extralegal factors matter. Legal and Criminologi- cal Psychology, 12, 21–39.

Paparozzi, M. A., & Gendreau, P. (2005). An intensive supervision program that worked: Service delivery, professional orientation, and organizational support- iveness. Prison Journal, 85, 445–466.

Quas, J. A., Bottoms, B. L., & Nunez, N. (Eds.) (2002). Child maltreatment and delinquency: Framing issues of causation and consequence. Children’s Ser- vices: Social Policy, Research, and Practice, 5(4), 245–248. Juvenile Offenders Abused as Children349 Salekin, R. T., Yff, R. M. A., Neumann, C. S., Leistico, A. R., & Zalot, A. A. (2002).

Juvenile transfer to adult courts: A look at the prototypes for dangerousness, sophistication-maturity, and amenability to treatment through a legal lens.

Psychology, Public Policy, and Law, 8, 373–410.

Sanborn, J. B., Jr. (1996). Factors perceived to affect delinquent dispositions in juve- nile court: Putting the sentencing decision into context. Crime & Delinquency, 42, 99–113.

Scannapieco, M., & Connell-Carrick, K. (2005). Focus on the first years: Correlates of substantiation of child maltreatment for families with children 0 to 4.

Children and Youth Services Review, 27, 1307–1323.

Sears, D. O. (1988). Symbolic racism. In P. A. Katz & D. A. Taylor (Eds.), Eliminat- ing racism: Profiles in controversy (pp. 53–84). New York: Plenum.

Shaver, K. G. (1985). The attribution of blame: Causality, responsibility, and blame- worthiness. New York: Springer-Verlag.

Snyder, H. N. (2004, September). Juvenile arrests 2002. Juvenile Justice Bulletin.

Washington, DC: Office of Juvenile Justice and Delinquency Prevention, U.S.

Department of Justice.

Stalans, L. J., & Diamond, S. S. (1990). Formation and change in lay evaluations of criminal sentencing: Misperception and discontent. Law and Human Behavior, 14, 199–214.

Stalans, L. J., & Henry G. T. (1994). Societal views of justice for adolescents accused of murder: Inconsistency between community sentiment and automatic legisla- tive transfers. Law and Human Behavior, 18, 675–696.

Stevenson, M. C., & Bottoms, B. L. (in press). Race shapes perceptions of juvenile offenders in criminal court. Journal of Applied Social Psychology.

Stevenson, M. C., Bottoms, B. L., Diamond, S. S., Najdowski, C., Stec, I., & Pimental, P. (2008, March). Jurors’ discussions of a defendants’ childhood maltreatment during capital sentencing deliberations. Paper presented at the meeting of the American Psychology Law Society, Jacksonville, FL.

Sweeney, L. T., & Haney, C. (1992). The influence of race on sentencing: A meta- analytic review of experimental studies. Behavioral Sciences and the Law, 10, 179–195.

Veltman, M. W. M., & Browne, K. D. (2001). Three decades of child maltreatment research: Implications for the school years. Trauma, Violence, & Abuse, 2, 215–239.

Vidal, S., & Skeem, J. L. (2007). Effect of psychopathy, abuse, and ethnicity on juve- nile probation officers’ decision-making and supervision strategies. Law and Human Behavior, 31, 479–498.

Weiner, B. (2006). Social motivation, justice, and the moral emotions: An attribu- tional approach. Mahwah, NJ: Lawrence Erlbaum.

Williams v. Taylor, 529 U.S. 362 (2000).

Widom, C. S., & Wilson, H. W. (in press). How victims become offenders. In B. L.

Bottoms, G. S. Goodman, & C. J. Najdowski (Eds.), Child victims, child offend- ers: Psychology and law. New York: Guilford Press.

Wordes, M., Bynum, T. S., & Corley, C. J. (1994). Locking up youth: The impact of race on detention decisions. Journal of Research in Crime and Delinquency, 31, 149–165.