Confrontation and the Re-Privatization of Domestic Violence
When the Supreme Court transformed the right of confrontation in Crawford v. Washington, the prosecution of domestic violence predictably suffered as a result. But commentators at the time did not anticipate how the Court’s subsequent Confrontation Clause cases would utterly misconceive the nature of domestic violence, producing a flawed understanding of what constitutes a “testimonial” statement. Although the Court’s definition was especially problematic in the domestic violence context, its overly rigid approach finally became intolerable in Michigan v. Bryant, a 2011 case that did not involve domestic violence. In Bryant, the Court resurrected a public–private divide that relegated domestic violence to quasicriminal status, at best. By distinguishing between “domestic” and “nondomestic” disputes and minimizing the harms and dangers associated with the former, the Court revived long-standing hierarchies that were ostensibly repudiated decades ago. In assessing the significance of the Crawford revolution after ten years, I focus here on this largely unremarked jurisprudential move, which raises the distinct possibility that a privatized notion of domestic violence infected the Court’s reasoning even before Bryant.
Since the Supreme Court revamped the Confrontation Clause, we have witnessed a rather jurisprudentially unsettled decade. A sound articulation of the meaning of “testimonial,” which Crawford introduced but failed adequately to define, continues to elude the Court, and lower courts have struggled to implement the testimonial concept.
Amid the Crawford confusion, properly treating statements made by crime victims to police officers has proved especially vexing. Two years after Crawford, in consolidated cases involving domestic violence, the Court offered a description of “testimonial.” In one of those cases, the victim’s on-scene statements to responding police officers were deemed properly excluded as testimonial, “part of an investigation into possibly criminal past conduct.”
Five years later, in 2011, the Court decided Bryant, where a gunshot victim spoke to police while lying in a gas-station parking lot. By a vote of 6–2, the Court applied a new multifactor test to allow into evidence the dying man’s statements, which were made in response to questions about “what had happened, who had shot him, and where the shooting had occurred.” This conversation with police officers “ended within 5 to 10 minutes.”
Critics assailed Bryant as a dramatic departure from the earlier cases, tangible proof that the Court was revising its perspective on Crawford. I have a different view of the case, and of the Confrontation Clause jurisprudence more generally. In my estimation, the problem was not Bryant. Or, to be more accurate, Bryant did not present the problem that others have identified. Instead, Bryant reflected the Court’s effort to remedy an earlier, unacknowledged failing: the testimonial definition was conceptually incoherent from the start.
Part I of this Essay contends that Bryant raised a set of compelling concerns—not for the integrity of Confrontation Clause jurisprudence, as others urge, but for the portrayal of domestic violence as a lesser crime. The Bryant Court retreated from its problematic definition of testimonial—which is most problematic in the domestic violence context—only by cordoning off domestic violence as an exceptional case that warranted the rejected formula. By using domestic violence to reify nondomestic violence as real crime, the Court resurrected the public–private divide that law-reform efforts had targeted with considerable success over time.
The notion that domestic violence is a crime against the state took hold in the 1970s, prompting a sea change in the way prosecutors handled cases that were formerly dismissed as private disputes. For the first time, the state was perceived as having a stake in holding batterers accountable, which meant that a case would sometimes proceed even without the victim’s cooperation. The transformed Confrontation Clause made “evidence-based prosecution,” as it became known, more difficult; Bryant further complicated this picture.
After discussing the implications of Bryant’s public–private dichotomy, Part II of this Essay considers the post-Crawford state of domestic violence prosecution. A brief conclusion follows, reflecting on the Court’s move toward re-privatizing domestic violence in the Confrontation Clause cases.
I. Bryant’s Resurrection of the Public–Private Divide
When Bryant was decided in 2011, the Court emphatically softened the testimonial definition articulated in Davis v. Washington. It seems that the Court had grown uncomfortable with the full implications of the transformed Confrontation Clause—although Justice Sotomayor, writing for the majority in Bryant, never conceded this point. What emerged from Bryant was a new multifactor test that purported to adhere to precedent while actually departing from it in important ways.
The new test was attacked for resembling the malleable reliability standard that Crawford overruled and for being generally indeterminate. But Bryant did not corrupt a conceptually sound jurisprudence; rather, it was born of incoherence. The Court’s definition of testimonial rests on the fallacy that exigency can be captured without reference to context—a misconception that would later contribute to the unraveling of the Court’s rigid approach to the testimonial question.The multifactor approach can best be understood, then, as a natural outgrowth of irresolvable tensions underlying Davis.
Lacking a principled framework, the Court can hardly be faulted for reverting to greater flexibility. This latest jurisprudential move is troubling, however, for reasons that have gone largely unexplored. In stunning ways, Bryant reconstructed a boundary between the public and private spheres that, for most of our history, worked to the profound detriment of battered women. Nearly fifty years of feminist law-reform efforts effectively challenged the entrenched belief that domestic violence, because it is private, lies outside the reach of state intervention. But in Bryant, the familiar public–private divide resurfaced.
Throughout the opinion, the Court underscored the public nature of the crime at issue in Bryant, where the victim, Anthony Covington, was shot by a drug-dealing acquaintance. In contrast, “purely private dispute[s]” were said to warrant their own special treatment. In the Court’s estimation, domestic violence cases are different from nondomestics, and they are different in ways that justify a constricted definition of exigency.
This contention rests on a number of misconceptions. At the outset, the Court posited that “[d]omestic violence cases like Davis and Hammon often have a narrower zone of potential victims than cases involving threats to public safety.” This claim is confusing, since it is unclear which cases “involv[e] threats to public safety;” surely not all nondomestic violence cases fall into this category. More important, the claim is also misleading, because it suggests that domestic violence cases do not tend to involve the potential for harm to secondary victims. As the Court candidly observed, “[b]ecause Davis and Hammon were domestic violence cases, we focused only on the threat to the victims and assessed the ongoing emergency from the perspective of whether there was a continuing threat to them.” But rather than remedy the earlier failing, the Court aggravated it by resorting to false dualisms.
Consider how the Court’s preoccupation with the public space delineated a private realm that, if not altogether lawless, was imagined as a place subject to a less-than-urgent, somewhat diluted, brand of law enforcement. By way of contrast, Covington met the police in a “public location,” an “exposed, public area,” thereby creating a danger to the “public safety” that, in no uncertain terms, trumped the unnamed domestic analogue. “Private safety” disappeared entirely as a concern.
Similarly, Bryant’s conception of public safety excluded children and other family members of the primary victim (again, responding police officers are at risk in all cases, although the Court contemplated only the danger from nondomestic calls). Likewise, just as the notion of public safety emphatically ignored family members, the Court’s “zone-of-potential-victims” analysis entirely overlooked the primary victim, who may herself remain a “potential victim,” particularly in the domestic violence context.
The Court also devalued the kinds of injury that tend to occur in private space. For example, it noted that “Hershel Hammon was armed only with his fists when he attacked his wife.” Unlike the situation in Bryant, which involved a gun, in Hammon “removing [the victim] to a separate room was sufficient to end the emergency.” This was one way of characterizing the incident in Hammon; the facts in the trial record, however, supported quite another. Here is what we know: police responded promptly to a reported domestic disturbance and found a “timid” and “frightened” woman; they also found a man who admitted to arguing with his wife but claimed—despite a living room in a state of “disarray” with “broken objects littering the floor” and “broken glass . . . in front of a glass heating unit that appeared to be broken with flames coming out the front of the unit”—that it never became physical. After police separated the two of them, the woman told police that her husband had thrown her into the shattered glass and punched her in the chest, and she said that she was in pain. Despite the efforts of police to keep the man away, he made “several attempts” to enter the room where the woman was speaking to an officer about the episode, and he became “angry” when the officer “insisted that [he] stay separated” from his wife “so that [the officers could] investigate what had happened.”
Perhaps, even on this telling, the Court would resist characterizing Hershel Hammon as “the perpetrator of a violent crime.” What is apparent, however, is that violence, as the Court conceived of it, must exceed a threshold before it even qualifies as an emergency. As Justice Scalia sputtered in dissent, “I do not look forward to resolving conflicts in the future over whether knives and poison are more like guns or fists for Confrontation Clause purposes.” This inquiry is exceedingly problematic, not just because cases will fall in the middle but because there is no good reason that the designation should matter to the ultimate question—that is, whether the statement was made during an ongoing emergency.
This is not to deny that Covington’s gunshot wound to the abdomen was more serious than the physical injury suffered by Amy Hammon—Covington died from his wound. But even so, Amy Hammon may have been crying for help and in desperate need of police protection in a way that Covington was not. Bryant missed this distinction, perceiving violence as necessarily entailing serious physical injury that occurs in a moment of time. Accordingly, the Court reduced to irrelevancies the characteristic features of domestic violence: ongoing patterns of coercion that escalate and thereby cause the victim to live in fear.
As deployed by Bryant, “the private” accomplishes much of the work of designating a statement as testimonial. Had Hershel Hammon fled by the time the police arrived, it is unlikely that his “on-the-loose” status would signal to the Court the existence of an ongoing emergency. Under Bryant’s logic, a man who had “only” used his fists against his wife would not be perceived as a danger to the public. This reading is further supported by the Court’s reminder that “ ‘a conversation which begins as an interrogation to determine the need for emergency assistance’ can ‘evolve into testimonial statements.’ ” And the Court provided additional guidance that is even more on point: “This evolution may occur if, for example, a declarant provides police with information that makes clear that what appeared to be an emergency is not or is no longer an emergency or that what appeared to be a public threat is actually a private dispute.”
In Bryant, private disputes became a construct with independent significance. These disputes, as the Court conceived of them, create their own, limited emergencies; they occur in private space, which shields the public from collateral damage; and they tend to inflict less injury than other crime. Private disputes are defined by their points of departure from violence inflicted by strangers, which is privileged as the paradigm of crime, properly subject to state regulation.
These boundaries are not new, which means that Bryant must be placed in sociohistorical context to appreciate fully its import.
The legal treatment of domestic violence originated in a right of chastisement that remained in effect until the late nineteenth century. Even after the formal chastisement right had been repudiated and wife beating had been criminalized, however, domestic violence remained widely accepted. In place of transparently hierarchal norms, the rhetoric of marital privacy developed to justify nonintervention in intimate relationships. Until recently, noninterventionist policies, and the privacy-based rationales used to support them, “saturated the criminal justice system at all levels—police, prosecutor and bench.”
This began to change in the 1970s, as challenges to police inaction resulted in improved training, significant procedural reforms, and substantially more domestic violence arrests. These developments, in turn, prompted a focus on prosecution. Professor Sack describes this evolution as follows:
As the number of domestic violence arrests started to increase, prosecutors began to examine new ways to handle these cases. They began to develop “no-drop” policies in which their decision to go forward on a domestic violence case was not determined by whether the victim wanted the case to proceed. Not only would this improve prosecution rates in domestic violence cases, but it would have a positive effect on arrests, as police are more likely to make an arrest if they believe the case will be prosecuted. In addition, as with mandatory arrest, by not making prosecution dependent on a victim’s decision to press charges, no-drop policies would reduce batterers’ attempts to intimidate or retaliate against victims to keep them from proceeding. . . . And, a no-drop policy would make it clear that the justice system takes domestic violence seriously and treats it as a crime. Domestic violence is a public safety issue that is not confined to the victim and offender, but impacts the community as a whole.
Introducing domestic violence as a public concern thus required implementing a host of tangible measures designed to ensure that such violence would be effectively prosecuted.
For present purposes, it is important to emphasize that the in-the-trenches treatment of domestic violence as a crime developed alongside deliberate efforts to dismantle a public–private divide that placed the latter off limits. This conceptual move, which must be counted among the great contributions of feminist legal theory, facilitated a legal progression toward equal protection for women in violent relationships.
From this perspective, Bryant is worse than poorly reasoned; in relying on a retrograde framework that—it seemed—had been largely dismantled, the case marks a regression. Perhaps this regression is outweighed, as a practical matter, by the greater flexibility now generally afforded the testimonial determination (although nondomestic cases are the ones that will most readily benefit from Bryant’s more contextual approach). More to the point, the Supreme Court was not compelled to resuscitate, then sacrifice, “the private” to achieve its desired end. Its willingness to do so speaks to the endurance of long-standing hierarchies.
II. Domestic Violence Prosecution After Crawford
The re-privatization of domestic violence has a less abstract meaning on the ground, where, soon after Crawford was decided, prosecutors began dismissing cases. Since then, courts have excluded as testimonial statements that would otherwise have been routinely admitted under the old reliability regime. And, at the earlier charging stage, prosecutors have undoubtedly taken into account the odds of losing on admissibility, although this measure of Crawford’s impact is quite difficult to gauge.
Whether Bryant will have a meaningful impact on domestic violence cases, and what that impact will be, is yet to be determined. Although the Court has now purported to make the existence of a relationship central to the testimonial analysis, a number of post-Davis cases classify on-scene statements by victims of domestic violence as nontestimonial, made in the course of an emergency. For a lower court so inclined to adopt this perspective, the Bryant Court’s multifactor approach—if not its artificial distinctions—may actually prove helpful.
A classification as testimonial need not doom a victim-absent prosecution, of course. Prosecutors interested in pursuing a case without a victim may be able to invoke the equitable doctrine of forfeiture. When the batterer’s efforts to control his victim resemble traditional witness-tampering methods—such as placing threatening calls from jail—there is no conceptual barrier to proving a defendant’s intent to procure the witness’s absence, a requirement the Court enunciated in Giles v. California. But to the extent that abusers employ tactics that depart from paradigmatic witness tampering, which often occurs, prosecutors may find it difficult to establish the requisite intent. This difficulty limits the applicability of the Court’s forfeiture framework to domestic violence, although the doctrine remains useful.
Overall, my sense is that the reworked Confrontation Clause has not doomed evidence-based prosecution but rather has curtailed it. In cases where prosecutors can no longer introduce a victim’s statements to police officers, the alternatives may now include forcing her to testify or dismissing the charges altogether. Efforts to hold abusers accountable have thus become noticeably more dependent on the participation of domestic violence victims.
As a result, one positive development has emerged: prosecutors are increasingly attuned to the myriad challenges that battered women face in assisting the state with investigation and trial. This reorientation has in turn produced a greater emphasis on providing the kind of support that can enable domestic violence victims to do the hard work of cooperating with the state. Ideally, prosecutors are adopting holistic measures aimed at addressing the full range of obstacles confronting the victim, whose participation in the criminal case often necessitates constructing a life away from the batterer. Best practices now include providing assistance that “support[s] the victim’s legal and non-legal needs, including housing, education, childcare and employment accommodations.” In essence, policies of this kind—which are by no means universal but are indeed widespread—deliberately target underlying causes of the familiar alliance between accuser and accused. By addressing the victim’s reasons for noncooperation, the state positions itself to proceed with the accuser, reconfiguring a triangular structure that has long plagued domestic violence prosecution.
Prosecuting domestic violence in a manner that empowers the victim is optimal, and recognizing victim support as integral to this model’s success should be celebrated. But these adjustments do not eliminate the state’s imperative to proceed sometimes despite the victim’s noncooperation. Cases where the victim’s expressed interests diverge from those of the prosecution are inevitable. These are the cases that squarely present the question of whether, as a normative proposition, charges should invariably be dropped simply because the victim does not wish to testify. In my view, the answer is no—that is, the decision must depend on the particular facts at issue. Although proceeding with a case may not always be the best prosecutorial response, the state has a powerful interest in redressing the crime (albeit, an interest that may on occasion be outweighed by other concerns). The public harm of domestic violence is real, notwithstanding a Confrontation Clause jurisprudence that disregards it.
Davis was fundamentally incompatible with the realities of domestic violence. Unmoored from context, the new approach to the testimonial inquiry posited an artificial ending to an ongoing emergency. Resolving an exigency often depends on an arrest and therefore on the victim’s recounting of past criminal conduct. By overlooking this fact, the Court dictated the exclusion of out-of-court statements that were functionally quite different from the paradigmatic Confrontation Clause violation. Yet this discrepancy went unacknowledged, and the Court failed to offer an alternative vision for the normative underpinnings of the confrontation right.
When the Court addressed the limits of the Davis approach, it implicitly acknowledged that context matters in assessing the contours of exigency. Bryant’s multifactor test conceded that an expansive list of factors was relevant to the analysis, but it did not elaborate on the norms underlying confrontation. Most surprisingly, the Court suggested that greater flexibility was warranted only with respect to nondomestic cases, leaving intact the discarded framework as applied to those cases least consistent with it.
The Court’s re-privatization of domestic violence in this manner raises the prospect that, in unspoken ways, the domestic aspects of Hammon and Davis (and, for that matter, Giles) were integral to their faulty analyses. The hierarchy of violence that surfaced in Bryant—and that indeed animated the Court’s reasoning throughout—may well have enabled the Court to tolerate, or even embrace, a model of confrontation that it would repudiate only when more public violence was at issue.
All told, a decade of Confrontation Clause jurisprudence suggests that advances in criminalizing domestic violence may be more tenuous than is generally believed.
* Professor of Law, Northwestern University School of Law. I am grateful to Sarah Buel, Bennett Capers, and Robert Mosteller for insightful comments on an earlier draft, and to Richard Friedman for his unwavering support of my work notwithstanding meaningful differences in our perspectives. This Essay is dedicated to the memory of Cheryl Hanna.
. 541 U.S. 36 (2004).
. 131 S. Ct. 1143 (2011).
. Id. at 1156.
. Crawford, 541 U.S. at 68.
. Davis v. Washington, 547 U.S. 813, 822 (2006).
. Id. at 829.
. Bryant, 131 S. Ct. at 1150.
. Id. at 1163.
. Id. at 1150.
. See, e.g., Michael D. Cicchini, Dead Again: The Latest Demise of the Confrontation Clause, 80 Fordham L. Rev. 1301 (2011).
. See infra notes 46–47 and accompanying text.
. See infra notes 49–50 and accompanying text.
. Bryant, 131 S. Ct. at 1153–54.
. Id. at 1174 (Scalia, J., dissenting) (“[T]oday’s decision is not only a gross distortion of the facts. It is a gross distortion of the law—a revisionist narrative in which reliability continues to guide our Confrontation Clause jurisprudence, at least where emergencies and faux emergencies are concerned.”).
. Id. at 1175 (Scalia, J., dissenting) (“The Court recedes from Crawford in a second significant way. It requires judges to conduct ‘open ended balancing tests’ and ‘amorphous, if not entirely subjective,’ inquiries into the totality of the circumstances bearing upon reliability . . . . This is no better than the nine-factor balancing test we rejected in Crawford . . . .”).
. See Deborah Tuerkheimer, Crawford’s Triangle: Domestic Violence and the Right of Confrontation, 85 N.C. L. Rev. 1, 20–32 (2006).
. Context is most relevant in the domestic violence setting, where the ongoing nature of abuse often means that police investigation and police protection are functionally the same. Apart from ignoring the continuing danger faced by the primary victim, decontextualized exigency determinations obscure the connection between domestic violence and imminent danger to children and police. Bryant acknowledged the centrality of potential secondary victims (“the first responders and public,” 131 S. Ct. at 1148) but only by implicitly denying that these victims exist in domestic violence cases and by ignoring the greatest risk of further harm by an at-large perpetrator—that is, the risk that confronts victims of domestic violence themselves. In an unfortunate irony, the cost of recognizing context has been adherence to the fiction of its absence in domestic violence cases, where context is uniquely essential to the meaning of exigency.
. Unlike Davis and Hammon,which “arose in the domestic violence context,” Bryant presented “a new context: a non-domestic dispute,” as the Court conceived of it. Id. at 1156.
. Id. at 1163–64.
. Id. at 1158.
. Id. at 1156.
. Id. at 1160.
. Id. at 1158.
. Tellingly, the Court was able to see this danger where the primary victim was not intimately involved with the perpetrator. The Court repeatedly worried that Covington was still in danger even after police had arrived at the scene: “Nothing Covington said to the police indicated that the cause of the shooting was a purely private dispute or that the threat from the shooter had ended. . . . What Covington did tell the officers was that he fled Bryant’s back porch, indicating that he perceived an ongoing threat.” Id. at 1163–64. There was “no indication that the shooter, having shot at him twice, would be satisfied that Covington was only wounded. In fact, Covington did not indicate any possible motive for the shooting, and thereby gave no reason to think that the shooter would not shoot again if he arrived on the scene.” Id. at 1166. According to Justice Scalia, the Court’s concern about further violence derived from an “active imagination” that “invent[ed] a world . . . where drug dealers hunt their shooting victim down and fire into a crowd of police officers to finish him off.” Id. at 1172 (Scalia, J., dissenting). Apart from whether this “dystopian view of Detroit,” id. (Scalia, J., dissenting), was warranted, the majority’s preoccupation with Covington’s safety is striking when juxtaposed with the Court’s utter disregard for the far greater possibility of ongoing violence in domestic cases.
. Id. at 1159.
. Hammon v. State, 809 N.E.2d 945, 947 (Ind. Ct. App. 2004), aff’d, 829 N.E.2d 444 (Ind. 2005), rev’d, 126 S. Ct. 2266 (2006).
. Hammon, 829 N.E.2d at 444.
. Id. at 498.
. Davis v. Washington, 547 U.S. 813, 820 (2006).
. Michigan v. Bryant, 131 S. Ct. 1143, 1159 (2011).
. Id. at 1176 (Scalia, J., dissenting).
. Id. at 1159.
. Id. (quoting Davis, 547 U.S. at 828).
. Bryant, 131 S. Ct. at 1159. I am suggesting that the private nature of a dispute—or, more helpfully, its source in a relationship of ongoing abuse—means quite the opposite of what the Court posited: in fact, past crime may reflect an ongoing emergency. To be clear, recognizing this point does not require accepting that the constitutionally relevant period of exigency extends indefinitely.
. Reva B. Siegel, “The Rule of Love”: Wife Beating as Prerogative and Privacy, 105 Yale L.J. 2117, 2122–23 (1996).
. Id. at 2130.
. Id. at 2150–74.
. Deborah Tuerkheimer, Recognizing and Remedying the Harm of Battering: A Call to Criminalize Domestic Violence, 94 J. Crim. L. & Criminology 959, 970 (2004) (internal citations omitted).
. See Joan Zorza, The Criminal Law of Misdemeanor Domestic Violence, 1970–1990, 83 J. Crim. L. & Criminology 46, 62 (1992).
. Emily J. Sack, Battered Women and the State: The Struggle for the Future of Domestic Violence Policy, 2004 Wis. L. Rev. 1657, 1672–73 (2004).
. See Cheryl Hanna, No Right to Choose: Mandated Victim Participation in Domestic Violence Prosecutions, 109 Harv. L. Rev. 1849, 1869 (1996).
. See Tom Lininger, Prosecuting Batterers After Crawford, 91 Va. L. Rev. 747 app. at 820 (2005) (76% of prosecutors responding to a survey of 60 prosecutors’ offices reported a higher dismissal rate after Crawford).
. See Robert K. Kry, Confrontation at a Crossroads: Crawford’s Seven-Year Itch, 6 Charleston L. Rev. 49, 52 (2011).
. See, e.g., People v. Lewis, No. A115107, 2007 WL 4206637, at *3–4 (Cal. Ct. App. Nov. 29, 2007); People v. Bradley, 862 N.E.2d 79, 81 (N.Y. 2006); State v. Shea, 965 A.2d 504, 510 (Vt. 2008).
. This is not to defend Bryant, which not only re-privatizes domestic violence but may also fail to protect the right of confrontation in cases where it should.
. 554 U.S. 353 (2008).
. See generally Sarah M. Buel, Putting Forfeiture to Work, 43 U.C. Davis L. Rev. 1295 (2010).
. See Wider Opportunities for Women, Prosecutor’s Guide to Safety and Economic Security for Victims of Violence Against Women 13 (2013), available at http://www.wowonline.org/wp-content/uploads/2013/10/WOW-ESS-Prosecutors-Sector-Guide.pdf [http://perma.cc/9WVP-5279].
. Id. at 9.
. Id. at 13.
. See Tuerkheimer, supra note 17, at 57 (identifying domestic violence as representing a deviation from the paradigmatic Confrontation Clause violation, which involves a particular configuration of relationships among accuser, state, and accused).
. Absent this reconfiguration, domestic violence cases tend to present a significant departure from the Confrontation Clause paradigm—accuser and state aligned against the accused—insofar as victims of abuse who resist cooperating with prosecutors ally themselves with the defendant against the state. Id.