Download Gender Inequality in Criminal Justice: Different Needs and Equal Protection and more Lecture notes Criminal Law in PDF only on Docsity!
THE FEMINIST CHALLENGE IN CRIMINAL LAW
STEPHEN J. SCHULHOFERt
INTRODUCTION
Feminist criticism of criminal law and criminal justice adminis-
tration has proliferated over the past decade and now touches
scores of doctrinal, practical, and theoretical issues. These critiques
and the associated proposals for reform are usually acknowledged
to be controversial (and even "radical") by proponents and
opponents alike. Yet, across a wide range of issues, the feminist
position has its basis in a simple fact that cannot be considered
debatable: criminal law is, from top to bottom, preoccupied with
male concerns and male perspectives.
In this Article, I explain why this seemingly tendentious claim is
not only accurate but uncontroversial. I then seek to show how the
male orientation of existing criminal law creates both the necessity
for reform and a major obstacle to doing it well.
The feminist challenge is to adapt male-oriented criminal
laws and practices to the concerns of a group of victims and
offenders who are normally left out of the picture. This turns
out to be difficult, and not just because of a lack of empathy
for the needs of women. Factoring female victims and female
offenders into the criminal law equation is hard because of many
conflicting concerns and commitments that most Americans share.
Three conflicts in particular will be central to the discussion that
follows.
First, although we want women to be treated the same as men,
sometimes equality cannot be achieved by treating two groups of
people the same way. We need to take differences into account.
t Julius Kreeger Professor of Law and Criminology and Director of the Center for Studies in Criminal Justice, University of Chicago. Portions of this Article were presented at McGeorge Law School in November 1994, at the Annual Meeting of the American Political Science Association in September 1994, and at the University of Chicago Law School in November 1993 as the Wilbur G. Katz Lecture. Research for this Article was supported by the RussellJ. Parsons Faculty Research Fund and the Sonnenschein Fund at the University of Chicago Law School. I have also benefited from the research assistance of Kimberly Dunn and Katharine Moir, and from the comments of Kathryn Abrams, Albert Alschuler, Mary Coombs, Anne Coughlin, Deborah Denno, Joshua Dressler, Leslie Goldstein, Elena Kagan, Dan Kahan, Jane Larson, Larry Lessig, Dorothy Roberts, Cass Sunstein, Michael Vitiello,Joan Williams, and Laurie Wohl.
2152 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
Yet drawing categorical distinctions between men and women undermines our ideals. This is the familiar debate concerning sameness versus difference. It pervades discussions of gender in other areas of the law (^) and discussions of equal treatment for racial minorities, the handicapped, and other groups. The debate plays out with some unexpected twists in criminal justice. A second dilemma is that we want to be sensitive to the nuances of context when gender issues are at stake. But effective protection of women also requires that women have clear-cut rights protected by clear rules. This is the old debate concerning rules versus discretion. Again, it plays out with some unexpected (^) twists in criminal justice. A third dilemma concerns the limits of theory. We need theory to help pinpoint the problems confronting women and to help organize thinking about solutions. But theory is not up to the task. Indeed, I will argue that in criminal justice, theory can never be equal to the task. Much of contemporary feminist discussion and scholarship center on developing new theories or on parsing the differences among theories and defending commitments to one of them or another. One theory emphasizes formally equal treatment.^2 A major competitor is a theory stressing the ways that culture and social practice subordinate women under laws that are formally neutral.' A third theory emphasizes context, caring, and connec- tion in lieu of what it views as a "male" commitment to abstract rights defined without regard to context. 4 One scholar has iden-
1 See, e.g., MARTHA MINOW, MAKING ALL THE DIFFERENCE: INCLUSION, EXCLUSION, AND AMERICAN LAW 21 (1990) ("The dilemma of difference grows from the ways in which this society assigns individuals to categories and, on that basis, determines whom to include in and whom to exclude from political, social, and economic activities."). 2 See, e.g., Wendy W. Williams, The Equality Crisis: Some Reflections on Culture, Courts, and Feminism, 7 WOMEN'S RTS. L. REP. 175, 175 (1982) (stating that courts should "rule that the privileges the law explicitly bestows on men must also be made available to women"); Wendy W. Williams, Notesfrom a First Generation, 1989 U. CHI. LEGAL F. 99, 99 (identifying the "'formal' equality" approach as "an insistence that laws not embody explicit sex-based distinctions"). 3 See, e.g., CATHARINE A. MACKINNON, FEMINISM (^) UNMODIFIED: DISCOURSES ON LIFE AND LAW 40 (1987) (discussing the "dominance approach" to the equality question); Christine A. Littleton, ReconstructingSexual Equality, 75 CAL. L. REV. 1279, 1282-84 (1987) (noting that courts routinely apply "phallocentric standards 'equally' to men and women's different reproductive biology or economic position to yield... unequal results for women"). 4 See, e.g., CAROL GILLIGAN, (^) IN A DIFFERENT VOICE: PSYCHOLOGICAL (^) THEORY AND
2154 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
search for a feminism that can guide reform in criminal justice and, perhaps, other areas. Despite the undoubted importance of theoretical insight, the most effective tools of reform at the present juncture are likely to be eclectic and atheoretical, and the most effective feminist scholarship is likely to be one that attends to the complexities of specific institutions and procedures. What is needed, I suggest, is a feminism of particulars, a recog- nition that real solutions are likely to lie deeply embedded in the details.
I. WHY CRIMINAL LAW IS "MALE"
The criminal justice system fits almost perfectly Lincoln's conception of a government of the people, by the people, and for the people. It fits perfectly, if you are willing to equate "the people" with the male half of the population. Criminal law is-and has been for centuries-a system of rules conceived and enforced by men, for men, and against men. There are counterexamples but not many. The law against pros- titution, which might make sense as a way to protect young and poor women from sexual exploitation, is not enforced that way. It is enforced almost exclusively against women.^8 The law notices women but prosecutes those it should be protecting. In many instances law enforcement does benefit and protect women. But overwhelmingly, criminal law is designed and implemented with men in mind. This observation should be considered completely uncontroversial. One way for the reader to confirm its accuracy is simply to pause for a moment and attempt to picture a typical offender. What does he look like? He is inevitably the subject of the inquiry. The criminal offender is disproportionately male, overwhelmingly so. In 1983, men and boys, 49% of the U.S.
(^8) See e.g., RICHARD SYMANSKI, THE IMMORAL LANDSCAPE: FEMALE PROSTITUTION
IN WESTERN SOCIETIES 88 (1981) (noting that although New York law was changed in 1964 to criminalize both male prostitution and the act of patronizinga prostitute, the effect of the change was minimal; in 1977, the number of males arrested was less than one-tenth the number of females arrested). Recent attention to the need for sanctioning prostitutes' customers has not substantially changed this picture. See Eleanor M. Miller et al., The United States, in PROSTITUTION: AN INTERNATIONAL HANDBOOK ON TRENDS, PROBLEMS, AND POLICIES 300, 313 (Nanette J. Davis ed., 1993) (noting the persistence of arrest patterns in which 70% of prostitution arrests are of females, while male customers account for only about 10% of arrests).
THE FEMINIST CHALLENGE
population, represented 78% of all property offenders and 89% of
all violent offenders.^
9
What about victims? The Women's Movement has taught us to
be far more aware of the victimization of women, and there is a
widespread sense that women are disproportionately victimized
by violence.^1 " There is an important truth in that perception,
but it is a complex truth. It is a truth that statistics seem to
contradict.
The victims of reported crime are disproportionately male,
again overwhelmingly so. Justice Department statistics indicate that
compared to women, men are 123% more likely to be the victims of
robbery and 161% more likely to be the victims of an aggravated
assault.1 1
One's immediate reaction, given widespread perceptions of
disproportionate victimization of women, is that statistics of this
kind must be distorted, and in part this is true. The great majority
of victims of domestic violence are female. 12 Outside of prisons
and other custodial institutions, over 90% of rape victims are
9 See BUREAU OF JUSTICE STATISTICS, U.S. DEP'T OF JUSTICE, REPORT TO THE NATION ON CRIME AND JUSTICE 41 (2d ed. 1988) [hereinafter REPORT TO THE NATION]. 10 See S. REP. No. 197, 102d Cong., 1st Sess. 33 (1991) ("Women bear the disproportionate burden of some of the most pernicious crimes, like rape, and some of the most persistent crimes, like beatings in the home."); S. REP. No. 545, 101st Cong., 2d Sess. 30-31 (1990) (noting "a spiralling 'gender gap' of violence... [in which] female victimization is increasing faster than male victimization (at least for some crimes)" (citations omitted)). " See BUREAU OF JUSTICE STATISTICS, U.S. DEP'T OF JUSTICE, CRIMINAL VICTIM- IZATION 1991, at 6 (1992) [hereinafter VICTIMIZATION IN 1991]. 12 In measuring domestic violence, a number of surveys suggest that male partners are victimized at about the same rate as female partners. See MURRAY A. STRAUS ET AL., BEHIND CLOSED DOORS: VIOLENCE IN THE AMERICAN FAMILY 266 (1980) (reporting a rate of 12.0 for husband-to-wife violence compared to a rate of 11.5 for wife-to-husband violence); Murray A. Straus & RichardJ. Gelles, Societal Change and Change in Family Violence from 1975 to 1985: As Revealed by Two National Surveys, 48 J. MARRIAGE & FAM. 465, 470 (1986) (reporting that in 1985, the rate of husband-to- wife violence was 113 per 1000 couples, while the rate of wife-to-husband violence was 121 per 1000 couples). But the overall data on the rates of assault by victim gender present a misleading picture because the assaults experienced by male victims tend to be concentrated disproportionately in the milder forms of slapping or hit- ting; the female partners are much more likely to suffer the most serious assaults. See Irene H. Frieze & Angela Browne, Violence in Marriage, in FAMILY VIOLENCE 163, 181 (Michael Tonry (^) & Norval Morris eds., 1989) (noting that "the average number of severely violent assaults by a husband against a nonviolent wife was three times greater than the average number of wives' assaults on nonviolent husbands").
THE FEMINIST CHALLENGE
an assumption of some contributory fault with regard to many of
the victims on the male side: victims who provoked fights,
assumed risks, or went looking for trouble. The victimization of
women seems disproportionate not simply to women's share of the
population but to their desert; the victimization of women seems
particularly unfair. The proper complaint, therefore, is not that
the victimization statistics lack "objectivity," because observa-
tions about women's disproportionate victimization involve an
irreducible normative element. That perceptions about women's
victimization rest in part on socially constructed value judg-
ments does not, of course, render such perceptions invalid or
unimportant.
What the data themselves suggest is that the criminal jus-
tice system's preoccupation with male offenders and male victims
is not exclusively an artifact of cultural bias in reporting
and charging behavior. In part, this preoccupation reflects
the nature of the underlying phenomenon of crime in our soci-
ety and probably most others. Nor would we expect (or want)
progress for women to increase women's participation as offend-
ers or as homicide, robbery, and assault victims.^19 So for the
foreseeable future, the raw material of criminal justice is likely
to remain, to an overwhelming extent, disproportionately male in
character.
Under these circumstances women have, until very recently,
remained at the margins of thought about criminaljustice problems.
The criminal justice system has been run by men, against men, and
for the benefit of men. Not so much different from the rest of
society, but more so.
The Sections that follow consider some of what needs
to be done to correct that imbalance in four areas of special
concern to women-domestic violence, rape, sentencing, and
prisons.
19 On the possibility that women's independence and increased workforce partici- pation might increase women's rates of offending, see infra text accompanying (^) notes 125-29. For a discussion of connections between women's independence and society's stereotypes of women as criminals, see Susan N. Herman, Thelma and Louise and Bonnie andJean: Images of Women as Criminals, 2 S. CAL. (^) REV. L. & WOMEN'S STUD. 53, 62 (1992); Elizabeth V. Spelman & Martha Minow, Outlaw Women: An Essay (^) on Thelma and Louise, 26 NEw ENG. L. REv. 1281, 1281 (1992).
2157
2158 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
II. DOMESTIC VIOLENCE
Lack of attention to domestic violence has a venerable tradition in the criminal justice system. Failure to prosecute abusive husbands was not just the result of inattention, because for centuries, the criminal law gave the husband an affirmative privilege to beat his wife in order to provide her with what was (^) seen as appropriate chastisement and instruction." The husband's formal privilege of chastisement was abolished by the end (^) of the nineteenth century,^21 but wife beating continued. The extent of it cannot be measured (^) precisely, but even the cautious studies point to high levels of abuse: 28% of all women experiencing a violent assault at some point in their marriages; 16% of married women assaulted by a spouse in^ a^ single^ year.
22
This culture of spousal abuse coexists with a police practice of not arresting wife beaters, especially in cases perceived as "ordinary" misdemeanor assaults. 2 1 Many episodes of domestic violence involve life-threatening attacks, assaults with guns, knives, (^) and other weapons, or brutal battering that leaves serious (^) physical injuries. Police are likely to take such cases more (^) seriously than the much
" See Anne M. Coughlin, Excusing Women, 82 CAL. L. REV. 1, 40-41 (1994) (noting that the husband's right to chastise his wife stemmed from "the belief that married women suffered from a volitional disability");Joyce E. McConnell, Beyond Metaphor: Battered Women, Involuntary Servitude and the Thirteenth Amendment, 4 YALE J.L. & FEMINISM 207, 232 (1992) (noting that "[t]he batterer's belief in a man's right to chastise his partner and to coerce sexual services remains entrenched in the law"). Blackstone attempted to rationalize the husband's privilege of chastisement by connectingit to the marital coercion doctrine, under which a husband could be liable for crimes the wife might commit in his presence (actual or constructive). It was only fitting, Blackstone argued, that the husband have a corresponding privilege of chastisement in order to deter his wife from actions that would expose him to criminal liability. See 1 WILLIAM BLACKSTONE, COMMENTARIES (^432) ("For, as he is to answer for her misbehaviour, the law thought it reasonable to intrust him with (^) this power of restraining her, by domestic chastisement, in the same moderation that a man is allowed to correct his servants or children .... "). " See, e.g., Fulgham v. State, 46 Ala. 143, 143 (1871) (abolishing a husband's right to chastise his wife); The Queen v.Jackson, (^) 1 Q.B. 671,681-82 (1891) (abolishing the right of marital chastisement in England); see also Bernadette D. Sewell, (^) Note, History of Abuse: Societa4 (^) Judicia, and Legislative Responses to the Problem of Wife Beating, 23 SUFFOLK U. L. REV. 983, 984 (1989) (noting that legal authorization of marital chastisement "continued to exist in many Western cultures until the late nineteenth century"). ' See Frieze & Browne, supra note 12, at 179. 2 See LAWRENCE W. SHERMAN, POLICING DOMESTIC VIOLENCE: EXPERIMENTS AND DILEMMAS 26-27 (1992) (describing (^) patterns of underenforcement of domestic violence laws).
2160 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
A. Mandatory Arrest Through the Lens of FeministJurisprudence
Domestic violence, a central issue in the struggle to assure dignity and equality for women, is extensively discussed in feminist literature.-" Yet there has been little (^) effort to consider how battles over theoretical commitments or attempts to refine theoretical paradigms (^) might inform efforts to begin solving this critical problem. One obvious irony involves the manner in which the tactic favored by many feminist reformers (mandatory arrest) collides with the part of feminist theory that advocates open-textured standards and attention to the nuances of relationships. Feminists inspired by the work of Carol Gilligan (^32) reject inflexible rights as a distinctively male conception. 3 Police officers who do not arrest battering husbands are not being very "male" in that sense. They are making ad hoc judgments, based on their intuitions about the whole situation. This sort of police discretion is presumably not what Gilligan really meant by an ethic of care and connection. Nevertheless, (^) the example highlights a more general point about the supposed clash between a male ethic of rights and (^) a female ethic of context and relationship. There is (^) nothing especially feminist about an ad hoc, discretionary approach (^) to making decisions. It all depends on the values that inform the exercise of the discretion. Nor should (^) we automatically count as "feminist" any ethic that emphasizes (^) caring, connection, and the continuity of relationships. There are situa- tions where that ethic is not only out of place, but dangerous to women. 34 Everything depends on the particular problem, the
arrest policies instituted in New York City). For references to the recent literature, see SANFORD H. KADISH & STEPHENJ. SCHULHOFER, CRIMINAL LAW AND ITS PROCESSEs 829 & n.2 (6th ed. 1995). 32 See, e.g., GILLIGAN, supra note 4. 3 See, (^) e.g., M. Kay Harris, Moving into (^) the New Millennium: Toward a (^) Feminist Vision ofJustice, (^) PRISONJ., Fall-Winter 1987, at 27,32 ("[In Gilligan's research] [m]en were more likely to employ a 'rights/justice' orientation [toward moral (^) issues] and women were more likely to reflect a 'care/response' orientation ... ."); Frances Heidensohn, Models ofJustice: PortiaorPersephone?Some (^) Thoughts on Equality, Fairness and Gender in the Field of CriminalJustice, 14 INT'L J. Soc. L. 287, (^) 295-96 (1986) (noting that men become "detached, autonomous and individualized" in their quest for justice, while women tend to focus on context and caring). ' An "ethic of care" poses related (^) dangers in such criminaljustice contexts as the juvenile court movement, civil commitment of the mentally ill, and rehabilitative models of sentencing. See StephenJ. Schulhofer, The GenderQuestion (^) in CriminalLaw, SOC. (^) PHIL. & POL'Y, Spring 1990, at 105, 124. In particular,
THE FEMINIST CHALLENGE
values presupposed, and the kinds of people who will make the
judgments. As Kathleen Daly observes:
When court officials define crime and impose sanctions, they use
relational reasoning and an ethic of care toward^ defendants^ ....
This female voice may not contain the same relational^ concerns
that women (or feminists) desire,^ but^ that^ is^ different^ from^ saying
that men's form of legal reasoning does not contain relational,
caretaking, or responsibility concerns. Thus, the problem in
criminal-court practices is not that the female voice is absent, but
that certain relations are presupposed, maintained, and
reproduced. 5
Scholars associated with what is often called "radical feminism"
approach this problem from the opposite direction. They^ argue
that women should have^ an^ absolute^ right^ to^ bodily^ integrity and
protection from aggression."^6 Formal^ equality (a^ "liberal^ feminist"
stance) is a poor benchmark here because men involved in fights
and minor assaults do not have the same need for state protection
from one another that women have for protection from assaultive
men. Thus, even if the police are equally inattentive to acquain-
tance assaults when victims are men or women, the state's seemingly
even-handed inaction is nonetheless an affirmative policy that
contributes to the subordination of women."^7 For feminists
committed to^ this^ view,^ society^ has^ an^ obligation^ to^ use^ arrest
powers vigorously when a man violates a woman's right to physical
safety.^3 " This is a coherent perspective, and surely an appealing
the model of caring and connection [is flawed] because ... conflicting interests are inherent in^ any^ criminal^ justice^ system^ that^ serves^ society's interests in deterrence and social protection. A system of^ criminal^ law premised on caring and connection will simply mask conflict and invite the abuses that vague standards of intervention have produced over and over in ostensibly benign programs. Id. (citation omitted). " Kathleen Daly, CriminalJusticeIdeologies and Practices in Different Voices: Some Feminist Questions About Justice, 17 INT'LJ. Soc. L. 1. 2 (1989). m See, e.g., MACKINNON, supra note 3, at 41 (noting that in domestic relationships, battery of women tends to be ignored, whereas "[w]hen [men] are hit, a person has been assaulted"). 17 See (^) id. at 76 (arguing that when government (^) refuses to prosecute for (^) marital rape, "the woman's obligation to deliver sexually is effectively enforced by the state").
' See Kathleen Waits, The Criminal Justice System's Response to Battering:
Understandingthe Problem, Forgingthe Solutions, in FEMINIST JURISPRUDENCE 188, 202 (PatriciaSmith ed., 1993)^ (arguing^ that^ "[a]rrest^ of^ the^ batterer^ is^ the^ central^ element of an effective police response"); Zorza, supra note 28, at 66 (arguing that in order to deter batterers effectively, more^ severe^ sanctions^ should be imposed).
1995]^2161
THE FEMINIST CHALLENGE
ments testing the effects of mandatory arrest were repeated, among
broader demographic groups and with^ better^ data-collection
procedures, in five other cities."' In three^ of^ them,^ arrest^ had^ a
greater deterrent effect than other responses only^ in^ the^ short^ run;
the effect tended to^ diminish^ over^ time,^ and^ within a^ year^ after^ the
initial intervention, suspects who had been arrested were more^ likely
to engage in repeat violence than those who had merely been
warned.^45 A^ reanalysis^ of^ the^ Minneapolis^ data^ revealed^ a^ similar
pattern in that city.^4 " Over time, in other words, arrest often
seems to have an "escalation effect," aggravating the subsequent
violence.
Advocates for battered women have attacked these more recent
studies, especially the "do nothing" strategy they seem to sup-
port. As one attorney writes, "We do not consider eliminating
arrest for... [robbery], [just] because it may not deter a particu-
lar... class of individuals."^4 " Unfortunately, that^ is^ a^ danger-
ous answer. The objective here is to protect battered women. If
arrest is not doing that, we need to consider using other solu-
tions, either in conjunction with arrest or as an alternative to it.
Moreover, the danger is not just that arrest may not deter, but that
in some situations, arrest may make matters worse. Mandatory
arrest may cause more violence to the very women^ we^ are^ trying^ to
help.
Other unintended harms to women have emerged as well. In
mandatory arrest jurisdictions, police are sometimes obliged to
arrest the abused woman because her partner alleges that she had hit
participation in the follow-ups, and the follow-up^ study^ was^ limited^ to^ only^ six months, a period when temporary deterrence^ effects^ may^ dominate^ to^ the^ extent^ that court proceedings remain^ possible.^ See^ Elliott,^ supra^ note^ 25,^ at^453 (noting^ that^ only 49% of the victims in the Minneapolis study completed all^12 follow-up^ interviews). 44 See SHERMAN, supra note 23, at 15-18 (reporting results from similar experi- ments conducted in Omaha, Charlotte, Milwaukee, Colorado Springs, and Miami). 45 See id. at 17, 188-87 (describing an initial deterrent effect of arrest, followed by a subsequent escalation in the likelihood of repeat violence in Omaha, Charlotte, and Milwaukee). 46 In Minneapolis, the deterrent effects of arrest (measured by victim interviews) decayed over time and disappeared after six months. See id. at^ 197.^ Although^ the Minneapolis study did not collect victim interview data after six months, the trend of the data suggests the possibility of an escalation effect after the six-month point. (When repeat violence was^ measured^ by^ official^ arrest^ records,^ arrest^ had^ a^ clear deterrent effect^ that^ continued^ for^ the^ entire^ 18-month^ period studied.^ See^ id.^ But official records may present a misleading picture^ of^ the^ actual^ level^ of^ battering,^ if arrest deters victims^ from^ reporting^ subsequent^ incidents^ to^ the^ police.) 47 Zorza, (^) supra note (^) 28, at 66.
1995]^2163
2164 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
him.^48 And there is some indication that visible, highly popular mandatory arrest programs have permitted legislators to reduce their support for more costly solutions like shelters for battered women.
49 Surprisingly, the tentative Minneapolis study and its recommen- dations for a more punitive approach received widespread attention and an immediately favorable reception, but public officials and the media have either attacked or ignored the more thorough studies that suggest the opposite conclusion." Theoretical^ and^ ideological commitments to punitive strategies and to a rights-oriented response to aggression seem to dominate any concern for designing operational programs which actually help abused women. Yet the best available evidence suggests that an across-the-board policy of mandatory arrest should be anathema^ to feminists.^ As^ Lawrence Sherman writes, using mandatory arrest to fight domestic violence "may make as much sense as fighting fire with gasoline." 51
C. Selective Mandatory Arrest?
An important qualification to this pessimistic assessment of mandatory arrest emerges when data from the follow-up studies are disaggregated. The escalation effect seems especially strong when the batterer is unemployed, but arrest does appear to have a net deterrent effect when the husband has ajob or other marks of^ social stability.^5 2 That finding triggers a further series of dilemmas. Should we arrest only the employed? If not, should we arrest in all cases, knowing that this will harm some women? A third possibility is to leave the decision to each officer on the spot. That would probably give us the worst of both worlds: police would arrest blacks and the underclass while^ letting^ off^ middle-class^ white
48 In Connecticut, mandatory arrest policies reportedly led to the arrest of^ both spouses in 14% of the cases. See Jan Hoffman, When Men Hit Women, N.Y. TIMEs, Feb. 16, 1992, (Magazine), at 23,26. Moreover, if children are present and no other caretaker for them is available, the children may have to be placed in state custody until one of the parents is released. See id. 49 See (^) SHERMAN, supra (^) note 23, at 255. 'o See id. at 135-36 (describing political attacks on empirical research and the unwillingness of the Wisconsin legislature to repeal its mandatory^ arrest^ statute^ after the Milwaukee study had indicated the harmful effects of that policy); id. at 266 (describing sparse press coverage of the Milwaukee findings); Sherman & Cohn, supra note 41, at 129 (stating that publicity about the Minneapolis study reached a large number of police departments and had a substantial influence on policy). 51 SHERMAN, supra note 23, at 210. 52 See id. at 155.
2166 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
D. Getting Tougher: Giving Content to "Arrest"
Instead of seeking to minimize arrest, especially for unemployed
husbands, a more promising approach might be to consider more
severe sanctions for all categories of offenders. Indeed, mandatory arrest may appear ineffective or dangerous in part because the response being advocated ("arrest") is mainly a slogan, not a fully specified policy. In many cities, suspects arrested on domestic violence charges are free within a few hours; among the six cities in which mandatory arrest experiments were conducted, average times in custody varied from twenty-four hours to only two.^5 " The constitutional right to bail" accounts for some of the rapid release times and may make that feature of the current landscape difficult to change. Nonetheless, rapid release times probably are not a major cause of the ineffectiveness of mandatory arrest, because most of the experiments show some deterrence over the short run, even when suspects spent very little time in jail.^59 The problem is that the deterrence effect of arrest tends to dissipate, and escalation effects begin to dominate, roughly six months after the arrest.^6 " The decay of deterrence effects could be slowed by making the initial arrest a more unpleasant experience, but that approach obviously risks making the escalation effects more severe as well. There is another reason why mandatory arrest is primarily a slogan, not a concrete policy. To advocate "arrest" says nothing about what should happen after the arrestee's inevitable release on recognizance or money bail. Husbands arrested for misdemeanor assault are almost never prosecuted.^6 ' An obvious option is to combine the increased use of arrest with a greater determination to get convictions and jail time. But the available data permit little confidence that prosecution provides the easy answer. In Milwau- kee and Charlotte, two cities in which mandatory arrest appeared to produce an escalation effect, the percentage of suspects prosecuted
57 See SHERMAN, supra note (^) 23, at 140-42. Longer times (^) in custody did not correlate cleanly with either the deterrence effects or the escalation effects, perhaps because times were poorly measured, and other factors that affect the severity of the arrest experience (jail conditions, for example) were not measured at all. See id. 59 "^ See^ U.S.^ CONST.^ amend.^ VIII. See SHERMAN, supra note 23, at 129, 141. 60 See id. at 189. 6" See Sherman, supra note 42, at 29 (reporting that in Minneapolis, only 4% of those arrested were ever convicted, and in Milwaukee, only 1% were ever convicted).
THE FEMINIST CHALLENGE
and convicted was only 1% and 28% respectively. 2 But substantial
escalation effects also occurred in Omaha, a city where 64% of the
arrestees were prosecuted and convicted.^6 "
Normatively, as well, the increased-prosecution approach poses
uncomfortable choices. Since employed offenders seem to be
deterred by arrest alone,' do we reserve prosecution and jail time
for the unemployed? That's not easy to live with. Or do we jail all
the offenders? If so, the men who had jobs often will lose them,
and middle-class wives who have sought help in the past may now
be afraid to call the police. And no matter which way we resolve
these issues, we still cannot be sure that a more punitive approach
will have a net deterrent effect overall. For the chronic, seriously
violent batterer, vigorous prosecution and substantial prison time
are usually appropriate. But unless we are willing to treat every
episode of domestic assault as a felony deserving a year or more in
prison, the deterrent and incapacitative effects of punishment, in
cases not involving a weapon or serious bodily injury, are inherently
limited.
E. Other Alternatives
Because mandatory arrest and other punitive responses have
such problematic effects, other options need to be explored. Yet
the array of promising alternatives is meager.
One seemingly constructive approach is for the officer on the
scene to refer the abuser to counseling services, possibly with the
added inducement of a court order or a threat to arrest after any
repeat episode. Unfortunately, evaluations of this option are just as
pessimistic as those concerning the use of arrest. There is as yet no
evidence that counseling reduces the offender's propensity for
repeat violence. 5 What is worse, because women are more likely
to remain with an abuser who is in counseling, the likelihood of
further victimization actually increases when the counseling option
is used.^66 Thus, counseling, like mandatory arrest, actually seems
62See SHERMAN, supra note 23, at 141-42. s See id.; Franklyn W. Dunford et al., The Role of Arrest in Domestic Assault: The Omaha Police Experiment, 28 CRIMINOLOGY 183, 193 (1990). " See supra text accompanying note 52. 6 See SHERMAN, supra note 23, at 249 (noting that "no randomized experiment yet demonstrat[es] (^) that court-ordered counseling can reduce the frequency or prevalence of repeat domestic violence"); Zorza, (^) supra note 28, at 67 (noting that "completing batterer treatment made no difference in stopping future violence"). 6See Zorza, supra note 28, at 67 (suggesting that victims may be "deceived" into
1995]^2167
THE FEMINIST CHALLENGE
one who has to leave her home? Why not require the abuser to
leave? Of course, the victim still can reject the offer; she could not
be required to leave. But when a police officer refuses to arrest the
abuser and encourages the victim to accept transportation to a
shelter, we can hardly view state policy as simply facilitating the
exercise of a wholly "voluntary" choice.
The Minneapolis experiment tested the option of requiring the
abuser to leave the premises (ostensibly for eight hours) without
arresting him.^7 " In reality, however, the abuser was free to return
as soon as the police left. Only arrest or an order of protection will
keep the abuser out of the home, but these responses can be
ineffective and dangerous, as we have seen. In any case, the
Minneapolis experiment showed that requiring the abuser merely to
leave produced high rates of repeat violence.^71 Against that
background, the fact that the victim has a right to remain on the
premises, while the abuser has none, can be dispositive only as a
matter of pure theory. When a terrorist is spraying the street with
bullets, those of us who have the right to remain on the sidewalk
may prefer not to stand our ground. In an imperfect world,
discretion may be the better part of valor. And the world of efforts
to control domestic violence is hedged by imperfections at every
turn.
F. Developing Effective Responses
The preceding considerations suggest that sharply etched,
theoretically satisfying positions are unlikely to carry us very far
toward effective solutions to the problem of domestic violence.
Helpful responses are more likely to be complex, eclectic, costly,
and elusive.
Several components of a constructive policy nonetheless can be
identified. First, complex and contingent response strategies must
not obscure the message that spousal abuse is a crime and that it will
not be tolerated. Every police response must be structured to send
the offender a clear warning that domestic violence is unacceptable
and that future incidents will be met with increasingly punitive
responses. Similarly, to combat any police tendency to take domes-
tic violence cases cavalierly, officers need guidelines that require an
energetic response. Mandatory arrest serves that function but only
7071 See See Sherman id. at 7, 12. & Berk, supra note 27, at 7, 9.
1995] 2169
2170 UNIVERSITY OF PENNSYLVANIA LAW REVIEW [Vol. 143: 2151
at some risk to the women we are trying to help. A more useful rule of thumb, therefore, would not insist on arrest per se, but mandatory action-action of some sort, from a list of strong, constructive alternatives,^7 2 and a mandatory report (normally (^) in writing), both to ensure a thoughtful response and (^) to inform officers who might be called to the same address (^) in the future. Specific response (^) options will necessarily reflect the services and programs (including prosecution programs) available in the particular jurisdiction; over the longer run, effort should focus on strengthening and evaluating alternative approaches. For chronic offenders and cases involving serious injuries, a vigorous prosecu- tion program should assure substantial jail or prison terms. In the most minor cases, a stern warning or a reference to counseling may be appropriate, especially if records are adequate to (^) assure that there have been no prior episodes of violence involving the same couple. For cases falling in between these extremes, arrest or some other means of separating the parties should be considered, with choices depending on the woman's desire to preserve the relation- ship, the history of past abuse, and other relevant circumstances. Where resources and back-up procedures permit, it may be appropriate to require the husband to leave and stay away long enough for the wife to obtain an effective order of protection. But if that avenue seems unlikely to afford real safety for the woman, (^) an offer to transport her to a shelter may be appropriate, if that is the course she prefers. The spousal-abuse problem illustrates two general points that apply to criminal justice and to probably most other areas of the legal system. First, legal, social, and institutional details are critical. Second, quick-fix statutes and doctrinal change do very little, and may even backfire, unless they are backed by changes in attitudes, personnel, and resources.
III. RAPE
I have just suggested that rigid rules will not work well in domestic violence cases. There, we need to rely on flexible standards. For rape statutes, I reach the opposite conclusion.
7 See SHERMAN, supra note 23, at 253-56 (listing options including transporting the victim to a shelter, taking the suspect or victim to a detoxification treatment center, and allowing the victim to decide if an immediate arrest should be made).
