Marital Separation and Lethal Male Partner Violence

Abstract

Findings reported by many researchers indicate that the association between marital separation and intimate partner femicide has achieved the status of a sociological empirical generalization. The primary objective of this article is to contribute toward the cumulative development of a conflict theoretic explanation of separation- associated femicide by creating and testing a deductive conflict resolution theory that explains the empirical generalization. The causal mechanism identified in the theory is the intensity of conflict that increases with participation in adversarial and separation and divorce proceedings. Interventions logically derived from the theory are presented in the penultimate segment. Limitations are identified in the concluding segment.

Keywords

separation conflict resolution collaborative/adversarial separation-divorce proceedings femicides

Introduction

Membership in the marital status categories of separation and divorce significantly increase the risk of an early or premature death (Sbarra, Law, & Portley, 2011). Compared with Canada and a number of European countries, the risk of a premature violent death is greater in the United States because rates of separation and divorce are far higher (Cherlin, 2009). Cherlin uses the metaphor of a spinning “marriage-go-round” carousel in which individuals jump on and off at a rapid rate to describe the situation in the United States. The rates for first, second, and third marriages are 54%, 67%, and 74%, respectively. One cause of the high rate of participation in separation and divorce proceedings and the life and death consequences associated with them is the implementation of no fault unilateral divorce laws in almost all states starting with Oklahoma in 1953.

Findings reported by a number of sociologists and economists indicate that one of the significant effects of the implementation of conflict-resolving no-fault unilateral divorce laws in the United States and Europe was a significant decrease in spousal femicides (husbands killing wives; Brassiolo, 2011; Dee, 2003; Dugan, Nagin, & Rosenfeld, 1999; Gillis, 1996; Stevenson & Wolfers, 1999). Economists identified the mechanism underlying the decrease as a “lowering of credible threat points” for leaving a relationship in which wives were being battered by husbands who may also have uttered death threats (Dee, 2003, p. 166). The mechanism, identified by sociologists Dugan et al. (1999), was “exposure reduction” (p. 187). Easier access to separation and divorce (plus the availability of domestic violence services such as shelters for battered women) enabled battered wives to leave the residence they shared with their batterers.

The mechanisms identified by these researchers apply to most wives who separated and divorced without the consent of their spouses. This article focuses on the few wives for whom exposure reduction and lowered threat points were not effective protective factors. Compared with the many husbands who were constrained by these mechanisms, a few were not. They settled conflicts associated with separation and divorce by killing wives who initiated such proceedings (Adler, 1999; Daly & Wilson,1995).

Most married couples attempt to settle conflicts associated with separation/divorce (primary residence of children, financial support, property division, collective debt) by participating in formal separation and divorce proceedings. Information about the effects of participation in these proceedings on intimate partner femicide were found to be “virtually non-existent” in the national criminal victimization and annual FBI Supplementary Homicide and Crime in the United States reports reviewed by Wardle (1994). This finding also applies to contemporary data sources such as the National Violent Death System reports that integrate information from a variety of sources, the national American Family Violence surveys conducted by Straus and Gelles (1990), and large-scale studies such as the Chicago Women’s Health Risk Study (Block, 2000). The gap in research on the effects of participation in separation and divorce on lethal male partner violence is fateful. One of the objectives of this article is to make a contribution toward closing it.

There does not appear to be as great a gap to close in research identifying separation as a risk factor for non-lethal and lethal male partner violence. The literature on violence against women is replete with publications that identify separation as a risk factor for intimate partner femicide or include separation as a risk factor in risk assessment instruments (Ellis, Sakinofsky, & Stuckless, 2012). The finding that femicide varies with recency of separation has achieved the status of an empirical generalization. Empirical generalizations are not explanations. They require explanation (Merton, 1957). Explanation often takes the form of identifying causal mechanisms that facilitate effective prevention (Conner, Duberstein, Conwell, Seidlitz, & Caine, 2001; Kraemer et al., 1997).

Major validation studies (Roehl, O’Sullivan, Webster, & Campbell, 2005) present evidence (in the form of coefficients) indicating that widely used risk assessment instruments such as the Revised Danger Assessment (RDA), DV-MOSAIC, Domestic Violence Screening Instrument (DVSI), and Kingston Screening Instrument for Domestic Violence (KSID) do a very good job of assessing the risk of serious violence against women. Evidence indicating that interventions derived from risk assessment scores and sub-scores in the instruments they validated actually prevented serious non-lethal and lethal violence against women in the sample was not presented.

The second objective of this article is to contribute to the cumulative development of a conflict resolution theory of separation-associated femicide by testing a deductive conflict resolution theory of femicides associated with participation in separation and divorce proceedings from which interventions aimed at prevention can be logically derived. To this end, we begin with a segment on conflict resolution dynamics.

Conflict Resolution Dynamics

In one of their earlier publications, Wilson and Daly (1989) describe the conflict resolution dynamic that sometimes results in femicide in the following terms:

Coercive use of violence by male partners is a potentially costly way of getting others to pursue (their) preferred agendas. Rather than making female partners wish to comply, violence inspires them to defy (male partner) perpetrators when the opportunity arises. Severe assaults can lead to severe self-defense measures. . . . It follows that violence is often the recourse of desperate people lacking the capacity to dispense positive incentives that would inspire more “voluntary” compliance with their male partner’s wishes. (p. 57)

In addition to self-defensive or retaliatory violence by female partners that leads to even more serious and occasionally lethal attacks by male partners, sub-lethal violence by male partners tends to increase the likelihood that their female partners will leave them. The imminent or planned departure of female partners tends to increase the intensity of conflicts between leavers and the left and consequently the risk of femicide.

Female partner resistance to the use of coercive control-motivated violence by male partners is prominent in the Wilson and Daly (1992) description of the conflict resolution dynamic in which conflict is settled by homicide. The modal conflict that is settled by the death of one of the partners is described in these terms:

Men . . . strive to control women, albeit with variable success; women struggle to resist coercion and to maintain their choices . . . and homicide . . . preceded by violent arguments . . . is an outcome of “slips in this dangerous game.” (p. 93)

The conflict resolution dynamics described by Gottman, Murray, Swanson, Tyson, and Swanson (2002) end in separation. Separation is the usual outcome when couples are involved in a relationship in which negative reciprocity has become an absorbed state. This relationship of “affective death” is characterized by frequent exchanges of contempt, criticism, defensiveness, and stonewalling. Negative reciprocity is a variable with some couples being more fully absorbed in this state than others. Negative reciprocity becomes an absorbed state when attempts made by intimate partners to settle conflicts (repair attempts) fail because difficulties in communication (negative sentiment override, fundamental attribution error) undermine them. Choices about participating in collaborative or adversarial legal proceedings aimed at settling conflicts associated with separation will tend to be influenced by the degree to which negative reciprocity as an absorbed state characterizes the relationship between separating partners.

Deductive Conflict Resolution Theory

The set of propositions constituting a conflict resolution theory of separation- associated femicide are stated below:

Proposition 1: The risk of femicide and femicide-suicide varies with the intensity of conflict between the separating parties.

Proposition 2: The intensity of conflict varies with the recency of separation and participation in proceedings aimed at settling separation-associated conflicts.

Proposition 3: Participation in collaborative proceedings decreases and participation in adversarial proceedings increases the intensity of conflict.

Proposition 4: Therefore, the risk of femicide will be greater among recently separated couples participating in adversarial proceedings.

Interrelations among the variables identified in the fourth proposition are presented graphically in Figure 1 and evidence supporting the theoretical assumptions follows.

Figure 1.

Femicide and femicide-suicide risk, intensity of conflict, and months since separation.

Findings

Proposition 1: The risk of femicide and femicide-suicide varies with the intensity of conflict between the separating parties.

Support for Aubert’s (1963) subjective definition of conflict as “feelings of mutual hostility” is indicated by the fact that “bad blood” has become a meme in North American culture (Ellis & Anderson, 2005). Bad blood existed between the Hatfield and McCoy families for some time before the alleged theft of a pig made the blood of both families boil. Killings of the Hatfields by the McCoys and the McCoys by the Hatfields followed. Findings reported by police officers completing homicide reports involving family members as perpetrators and victims indicate that increases in the temperature of the blood (intensity of conflict) to the point where arguments are used to try to settle them are frequently followed by the use of homicide as a conflict resolution tactic). Specifically, the motive of “argument-escalation” was reported in over half (53%) of the 116 spousal homicides investigated by Kowalski (2005, p. 55), Silverman and Kennedy (1993, p. 71), and Wilson and Daly (1994, p. 5). Theoretically, the association between intensity of conflict and lethal violence had been theorized by Black (2004).

Findings reported by Gillespie, Hearn, and Silverman (1998) indicate that the odds of suicide following homicide are highest when “the murderer was intimately involved with his (female) victim” as lover or husband (p. 55). These researchers did not include separation or symmetry/asymmetry in emotional attachment among the 19 independent variables whose coefficients were reported in Table 1. Had they included them and used the intensity of conflict as a control variable in a multivariate analysis, their findings may have supported the following proposition: The greater and more extreme the emotional attachment of a male to a female partner who decides to end the relationship, the greater the intensity of the conflict between them and the greater the likelihood that lethal violence will be used by him to settle conflicts associated with her decision.

Table 1.

Relationships Among NRAS, Intensity of Conflict, Ordering, and Type of Conflict Resolution Proceeding.

NRAS	Intensity of conflict	Ordering	Proceedings
Fully absorbed	Very high	Public	Adversarial
Highly absorbed	High	Private	Adversarial
Partially absorbed	Moderate	Private	Adversarial or collaborative
Minimally absorbed^a	Very low	Private	Collaborative

Note. NRAS= negative reciprocity as an absorbed state.

Includes couples who settle disagreements and low intensity conflicts on their own and couples who settle them with the assistance of third parties.

In many cases of separation-associated murders characterized by the presence of extreme emotional attachment or dependency, the intensity of conflict over separation is so high as to overcome norms internalized by male partners proscribing killing anyone but especially their wives (Daly & Wilson, 1988; Flynn et al., 2009; Milroy, 1993). Occasionally, “passionate rage” results in the “over-killing” of female partners who leave or plan to leave their partners and/or who are or are suspected of being sexually unfaithful (Crawford & Gartner, 1992, pp. 45-46; Wolfgang, 1958, p. 55).

Proposition 2: The intensity of conflict varies with the recency of separation and participation in proceedings aimed at settling separation-associated conflicts.

Findings reported by Ellis, Stuckless, and Smith (2014) indicate that a high proportion of femicides occur shortly after separation. The authors report findings indicating that on average, 47% of 236 femicides occur within 3 months and 74% of them occur between 2-6 months after separation. In the segment that follows, the period between 1-6 months following separation is referred to as the triadic phase of separation.

Separation has at least two phases. The first is a dyadic (“self-help”) phase in which the parties attempt to settle conflicts associated with separating themselves while they are still living together. In the second triadic (“third party involved”) phase the parties are living in separate residences and are participating or about to participate in formal separation proceedings. The intensity of conflict can increase to the point where femicide is used as a conflict resolution tactic in both phases (Goussinsky & Yassour- Borochowitz, 2012), but it more likely to reach its apogee in the second phase for two reasons.

First, in the separating phase, when the parties are still living together, there may be some uncertainty on the part of one or both parties as to whether the current separation is permanent, especially if she has left and returned on one or more previous occasions. In the second “third party involved” phase, uncertainty is replaced by certainty, especially when the partner who has left initiates formal separation proceedings. Separation is associated with significant emotional arousal generally, and hostility in particular, in male partners who see themselves as being “dumped forever” by their female partners (Arendell, 1995; Buss, 2005; Coser, 1956; Simmel, 1955; Weiss, 1975).

Second, the failure of intimate partners to settle separation-associated conflicts themselves tends to escalate the intensity of conflict.

During the triadic phase, conflicts associated with separating and divorcing can be settled at trial. Conflicts settled by trial are referred to as being “publicly ordered.” The vast majority—over 95%—of separations and divorces are “privately ordered” (Mnookin & Kornhauser, 1979). Private ordering means settling conflicts in the shadow of family law and family courts but not in the presence of judges who adjudicate outcomes of trials that are open to the public. In their frequently cited article, Mnookin and Kornhauser (1979) identify only one level of private ordering. I find it useful to differentiate among levels and types of private ordering.

Excluding parties who separate by simply leaving for places unknown and parties who do not contest separation, separating parties can settle conflicts associated with separation themselves by engaging in collaborative negotiations resulting in “consent agreements” filed in court. We shall refer to this as “private ordering—Level 1.” Private ordering agreements at this level are most likely to be achieved by parties who were involved in relationships characterized by minimal absorption in negative reciprocity as an absorbed state. Parties, especially those with children, who cannot negotiate a separation agreement on their own, usually invoke the assistance of third parties. We shall refer to this as “private ordering—Level 2.”

Third parties providing assistance include divorce mediators and lawyers. Some lawyers practice collaborative family law, others engage in problem-solving positional bargaining, and some engage in adversarial positional bargaining. My theoretic assumption is that the choice of triadic proceeding by separating partners will be influenced by their relationship history. Recall that negative reciprocity as an absorbed state is not a category but a variable. The association between relationships that vary in the degree to which negative reciprocity has become an absorbed state and participation in publicly/privately ordered adversarial and collaborative proceedings is described in Table 1. Separating partners who experienced mainly relationship-ending disagreements they could not settle themselves are included in the last group. Disagreements are differentiated from conflict by the absence of mutual feelings of hostility (Ellis & Anderson, 2005).

My theoretic assumption is that partners involved in “high intensity conflict relationships” will choose to participate in proceedings that greatly increase the intensity of conflict and those involved in very low intensity conflicts will choose to participate in proceedings that decrease the intensity of conflict. Many of the separating partners in the highly and partially absorbed groups will choose to participate in proceedings that increase the intensity of conflict to a variable degree, and some of them will choose to participate in proceedings that either do not increase the intensity of conflict or decrease it. Here, it is important to note the following specification: The effect of conflict intensity at any level on the use of lethal violence to settle conflicts can be moderated by the choice of proceeding (Cummings & Davies, 1994).

Relationships among the four factors referred to here are described in Table 1.

Proposition 3: Participation in collaborative proceedings decreases and participation in adversarial proceedings increases the intensity of conflict.

Adversarial proceedings in family courts may be rank-ordered with respect to the degree to which participation in them increases the intensity of conflict. The frequently cited statistic that less than 5% of separations and divorces “go to court” seriously underestimates the proportion of publicly ordered settlements because a far greater number of couples appear in family courts on contested motions (mainly parenting and financial support matters) after they separate or divorce. Participation in proceedings that are publicly ordered—litigated and re-litigated in open family courts—is ranked first with respect to the escalation of conflict.

Negotiations conducted by adversarial family lawyers are ranked second. Findings reported by Pruett and Jackson (1999) and Rimelspach (2001) indicate that representation by adversarial lawyers significantly increases the intensity of conflict. The literature reviewed by Ellis et al. (2012) indicates that the escalation of conflict occurs for one or more of the following reasons:

Participants use adversarial proceedings to harass, threaten, and abuse each other (Arendell, 1995; Canadian Panel on Violence Against Women, 1993; Durham Response to Woman Abuse, 2007; Pruett & Jackson, 1999);

Transaction costs–psychological stress, frustration, anxiety, delay, and financial–are high, sometimes very high (Ellis & Stuckless, 1996; Mnookin, 2010; Sander & Goldberg, 1994);

When children are involved, parenting and financial support arrangements imposed by judges are perceived to be zero-sum and morally outrageous by fathers who view them as the outcome of collusion between judges and mothers (Arendell, 1995; Braver & O’Connell, 1998; Kurz, 1995);

Imposed outcomes (minutes of settlement, court orders, and judgments) are perceived to be unfair by one or both parties for a number of reasons: One partner could not afford to pay for a lawyer, or to pay for one as experienced and effective as the one hired by the partner with greater financial resources; both parties represented themselves, but one of the parties was more articulate and knowledgeable about relevant family law, documents, and court procedures than the other and therefore presented the merits of her case far more effectively; judges were perceived to be biased against fathers (Arendell, 1995; Braver & O’Connell, 1998; Durham Response to Woman Abuse, 2007);

The submission of affidavits that include exaggerated, embarrassing, hurtful, and false allegations (Ellis, 1994);

In cases where one of the parties has a lot to gain or lose financially and/or romantically, strong motivation to avoid “a costly, dragged out divorce” increases the risk of homicide (Adams, 2007; Canadian Press, 2012).

The contribution made by the adversarial family law system and adversarial lawyers toward increasing the intensity of conflict is described in findings reported by Spanier and Casto (1979). They selected a sample of 28 males and 22 females who recently divorced and who agreed to be interviewed. Reports by a number of respondents, such as being “forced further apart” by their participation in the family law system and being encouraged by their lawyers to use a variety of “dirty tricks,” led Spanier and Casto to conclude that family lawyers and the adversarial family law system “appeared to encourage couples to become adversaries to a greater degree than they already are . . . to aggravate relations with the spouse . . . and to upset and humiliate people” (pp. 214-215).

In a survey conducted by Hedeem and Salem (2006), adversarial lawyers represented a fairly large proportion of the 219 lawyers included in their sample. Schneider and Mills (2006) analyzed data from their survey, and this was one of their major findings: “compared with civil, criminal, commercial, corporate, property and other lawyers, family lawyers had the highest percentage of unethically adversarial lawyers” (p. 617). Ethical adversarial and unethical adversarial lawyers accounted for almost 40% of the family lawyers in the sample. The corresponding figure for all lawyers in the sample is 33%. These findings led the authors to conclude that adversarial lawyers generally, but unethical adversarial lawyers in particular, were “engaging in . . . behavior that destroys . . . relationship(s)” beyond the damage done to relationships by the parties themselves. The relationship damage caused by the adversarial lawyers representing a divorcing couple (Brenda and Thomas) is described by Mnookin (2010) in these terms.

Hostility between Brenda and Thomas bordered on demonization and could lead to all- out warfare . . . worse yet, their lawyers seemed to be egging them on . . . exchanged half a dozen venomous letters without managing to resolve the most urgent and basic issues of this early phase. (p. 213)

Findings reported by a number of researchers (Arendell, 1995; Durham Response to Woman Abuse, 2007; Pruett & Jackson, 1999) provide support for the theoretic assumption that Public Ordering (“going to court”) increases the intensity of conflict beyond the baseline level of intensity that separating/divorcing parties bring with them to family court. Findings reported by ver Steeg indicate that couples do not actually have to go to court for the intensity of conflict between them to be elevated because the “shadow” cast by this possibility motivates family lawyers to “prepare and process all cases as if they were going to court” (2003, p. 231).

Participation in collaborative proceedings such as mediation tends to not only decrease the intensity of conflict below the level brought to them by separating parties but also to improve the quality of post-separation relationships between them (Bennett & Hughes, 2005; Ellis & Anderson, 2005; Emery, 2006; Emery et al, 2001; Fine, 2001; Kelly, Gigy, & Hausman, 1988; Pearson & Thoennes, 1984). Research and reviews of the literature conducted by Ellis and Stuckless (1996) and Ellis et al. (2012) indicate that divorce mediation may achieve these outcomes for one or more of the following reasons:

Participation is governed by collaborative norms;

Outcomes are determined by the parties themselves;

Outcomes on many or most outcomes are more likely to be perceived as positive sum;

Values and cognitions underlying conflicts are more likely to be discovered and reconciled;

Mediators facilitate constructive communications and problem solving, neutralize and reframe destructive communications;

Mediated agreements are likely to be perceived as fair by the parties when mediators (a) deny more powerful parties the opportunity to coerce less powerful parties by engaging in shuttle mediation where the parties do not meet face-to-face or communicate directly with each other and (b) use effective power-balancing interventions derived from a process theory of interpersonal power when they facilitate mediations that may or may not be face-to-face.

Based on interviews with a non-probability (convenience) sample of 75 divorced fathers, Arendell (1995) found that a significant proportion (40%) “threatened explicitly or resorted to acts of violence against the former wives since the end of their marriages” (p. 118). The alleged cause of their conflict instigated violence was perceived victimization by the adversarial family court system and their “intense bitterness” toward a system that escalated and maintained a “war without end” between divorcing fathers and mothers (Chapter 4). Five of the 75 men in Arendell’s sample participated in divorce mediation. They “cooperated during the divorcing process, agreed upon desired outcomes and continued to collaborate in parenting subsequent to divorce.” No violence against female ex-partners was reported for them (p. 76).

Findings reported by a number of other researchers indicate that the use of collaborative methods of settling conflicts has positive effects on relationships between intimate partners and their children that are as robust as the negative effects of using adversarial methods of attempting to settle them (Cummings & Davies, 1994; Fine, 2001; Odoms, 2001; Rogge & Bradbury, 1999; Scanzoni, 1982).

Proposition 4: Therefore, the risk of femicide will be greater among recently separated couples participating in adversarial proceedings.

Longitudinal aggregate data (annual rates of separation and divorce and homicide) collected from official French sources (e.g., Annuaire statistique de la France) and motives identified in coroner reports were analyzed by Gillis (1996) using a multivariate statistical method of analysis appropriate for analyzing time series data. The following two findings are noteworthy. First, the rate of participation in adversarial separation and divorce proceedings was a significant predictor of adultery- motivated . . . domestic homicides by males. Second, the rate of participation in adversarial separation and divorce proceedings is positively associated with “spontaneous femicides and homicides” and negatively associated with “premeditated domestic homicides” (p. 1273).

During the 1960s and 1970s, an increasing number of U.S. states with mutual consent divorce regimes (32 by 1978) adopted divorce regimes that permitted divorce without the consent of the other party. Dee (2003) investigated how unilateral divorce laws influenced spousal homicides. To this end, he collected information on spousal homicides and femicides included in the FBI’s Supplementary Homicide Reports for the years 1968-1978 and how “marital property was treated in divorce settlements” in different states with the same unilateral divorce regime (pp. 167-169). In common-law states, marital property or assets were distributed in a way that favored husbands. In community-property states, marital property was distributed in a way that favored wives. The major finding reported by Dee was that spousal homicides increased by over 20% in the four common-law states with easy access to divorce proceedings and marital property distribution that favored husbands.

The mechanism underlying this increase in these states was, we hypothesized, an increase in mutual feelings of hostility associated with participation in adversarial proceedings plus increased hostility toward husbands caused by the impoverishment of wives. Dee found no unilateral divorce regime states with marital property distribution that favored wives, but findings reported by Arendell (1995) indicate that the same mechanism underlies spousal femicides by male partners who see themselves as being emasculated, impoverished, and/or stripped of their rights by decisions made by family court judges adjudicating adversarial proceedings.

Logan et al. (2008) collected data on 408 homicide-suicide incidents reported in National Violent Death Reporting System reports from 17 states covering the years 2003-2005. Of the 191 perpetrators of femicide-suicide who used a gun to kill the female partners, 57% (n = 109) were involved in adversarial family court proceedings or were living with partners who intended to initiate them. The modal context for femicide-suicides was “intimate partner conflict” and current participation in adversarial separation and divorce proceedings were found to be “highly common amongst perpetrators of femicide-suicide” (p. 1060). A number of theorists and researchers (e.g., Easteal, 1993) conceive of femicide-suicides as femicides.

High conflict, young low-income couples were randomly assigned to a collaborative proceeding (mediation for an average of 5 hr) and an adversarial custody settlement proceeding by Emery, Laumann-Billings, Waldron, Sbarra, and Dillon (2001). Of the 71 families originally assigned to collaborative and adversarial groups, 27 mothers and 25 fathers in the mediation group and 25 of the mothers and 23 of the fathers in the adversarial group—hearings before a judge—were followed during the entire 12-year follow-up period. Four findings are noteworthy.

First, residential parents (parents whose homes were the primary residence of the children) in the mediation group gave better “grades” to non-residential parents in all 10 areas of parenting that were measured. Second, the process of decision making improved significantly. Indicators of “improved significantly” included working collaboratively as co-parents, respect for the voice of the other party, and adopting a longer run perspective. Third, fathers in the mediation group were more satisfied with the process and outcomes of mediation than fathers in the litigation group and the difference was statistically significant. Fourth, levels of conflict reported by couples in the mediation group were lower even though opportunities for conflict were far greater in the mediation group because couples in this group reported significantly greater parental involvement and contact with each other and their children than parents in the adversarial group. For couples who participated in mediation, these findings are entirely inconsistent with use of violence to settle conflicts and no violence was reported.

In a quasi-experimental field study published in 1984, Pearson and Thoennes reported findings similar to those reported by Emery et al. (2001). Compared with couples in the litigation group, couples participating in divorce mediation reported more positive short and longer term effects such as “improving the relationship” and “lower re-litigation rates” that would tend to decrease the likelihood of intimate partner violence.

Contributions

Five contributions come to mind. First and possibly for the first time, an empirically grounded, well established isolated proposition linking separation with femicide was derived from a deductive sociological theory. In Merton’s (1957) influential discussion of sociological theory (pp. 96-97), this is the first step in transforming an atheoretic empirical generalization linking a risk factor (separation) with femicide into a social scientific law, albeit one that may not be applicable in “classic patriarchies” (Kandiyoti, 1988) such as some regions in northern India today and in England during the Middle Ages up to 1530. These regions and this society “were neither separating nor divorcing locations and (in them) death was virtually the sole agent for dissolving marriage” (Singh, 2012; Stone, 1993).

Second, practitioners can logically derive at least two effective interventions aimed at prevention from a deductive theory that identifies causal mechanisms underlying an empirical generalization. First, as the first few months following separation are also the period during which separation and divorce proceedings are initiated, interventions aimed at prevention should be concentrated on recently separated female partners.

Third, the family law system should be changed in the direction of making voluntary participation in collaborative proceedings usual, and participation in adversarial proceedings residual. By residual, I mean reserved for cases in which one or both parties refuse to disclose financial information, or which require precedent-setting judgments. The Mandatory Information Program (MIP) recently implemented in Ontario provides an opportunity for separating and divorcing couples to make informed decisions about their choice of proceeding. Reasons for choosing mediation and collaborative lawyer representation are included in the presentations made by lawyer, mediator, and mental health presenters. Participation by Applicants and Respondents is mandatory in the sense that they cannot take a further step in legally processing their cases without obtaining a certificate of attendance at an MIP session (Attorney General of Ontario, 2010).

Ontario’s MIPs do not provide estranged participants with a theory/research informed handbook about and/or expose them to a video on constructive conflict resolution that is defined as a process that includes decreasing or at least not increasing the intensity of conflict. The conflict resolution theory described here suggests that their inclusion in MIPs may reduce reliance on the use of violent and abusive conflict settlement tactics during and following participation in separation and divorce proceedings.

Fourth, a deductive theory of separation-associated femicide provides a “rationale . . . (that offers) . . . a ground for prediction which is more secure than mere extrapolation from previously observed (risk factors)” (Merton, 1957, p. 98).

Fifth, a conflict resolution theory of separation-associated femicide would appear to be called for by the persistence and escalation of conflict during and following separation. For example, findings reported by the Centers for Disease Control and Prevention (CDC) indicate that “relationship conflicts play a strong role in the roughly 50,000 violent deaths in the U.S., each year (and) new information suggests that those conflicts may be more important than previously thought” (Barker, 2006:48). Buchanan and Heiges (2001) report findings indicating that “over two-thirds of couples who separate experience high conflict during the process.” Based on her study of homicide using FBI Supplementary Homicide Report data, Jensen (2001) concluded that “conflicts and quarrels (were) the most common motive for women’s homicide” (p. 47). CDC researcher Joe Logan (personal communication, August 5, 2013) reported that “Narrative data in the National Violence Death Reporting System commonly state that marital discord and even legal proceedings around divorce and separation and/or custody of children were common precipitators of femicide-suicide as perceived by witnesses and/or other informants.”

Limitations

The conflict theoretic explanation of femicides following participation in adversarial separation proceedings can be challenged by a social selection explanation hypothesis, which states that separated couples constitute a “relatively violent subset of all couples . . . with a history of discord” (Wilson & Daly, 1994, p. 8). If these couples are concentrated among those who are most likely to choose to participate in adversarial separation proceedings, then selection may be “incidental” to separation-associated femicides via their effect on choice of proceedings. However, qualitative data from cases analyzed by Wilson and Daly (1993) indicated that the association between separation and femicide was “more than incidental” (p. 8). This evidence undermines, but does not eliminate a selection explanation.

A selection explanation can be eliminated by implementing a field experimental study design with random assignment of thousands of separating couples to collaborative and adversarial proceedings and following them for a number of years. Such a study is neither ethical nor feasible.

Findings reported by researchers using different non-field experimental study designs do not support a selection explanation. For example, Mahoney (1991) and Dobash, Dobash, Cavanagh, and Medina-Ariza (2007) report findings indicating that some femicides associated with separation involved conventional non-violent couples as victims and perpetrators with femicide being a first-time violent criminal offense. These couples do not constitute a violent sub-set of separating couples.

In the experimental field study conducted by Emery et al. (2001), couples constituting a probable violent sub-set (young, low income, high conflict) were randomly assigned to adversarial and collaborative (mediation) proceedings and a significantly higher proportion of the mediation group couples exhibited behavior inconsistent with intimate partner violence. Still, I cannot claim to have eliminated the contribution made by selection to a conflict theoretic explanation, which specifies intensity of conflict as a causal mechanism underlying the association between recent separations and femicides. In fact, the contribution made by selection is acknowledged in the specification of an association between couple variations in negative reciprocity as an absorbed state and choice of proceeding.

Second, the evidence presented in support of a conflict theoretic explanation is persuasive but not compelling. One reason for this limitation is the extreme paucity of data on participation in separation and divorce proceedings. Perhaps these data will be included in future narrative reports published by the National Death Reporting System.

Finally, findings from non-probability samples reported here may be valid but not generalizable to the population of separating/divorcing couples. More accurately, findings can be generalized, but error cannot be statistically estimated.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biography

Desmond Ellis is based at York University’s LaMarsh Centre for Child and Youth Research and is currently involved in completing a book, Marital Separation and Lethal Domestic Violence: Theory, Research, Prevention, being published by Elsevier/Anderson.

References

Adams

(2007). Why do they kill? Men who murder their intimate partners. Nashville, TN: Vanderbilt University Press.

Adler

(1999). “If we can’t live in peace, we might as well die”: Homicide-suicide in Chicago. Journal of Urban History, 26, 3-21.

Arendell

(1995). Fathers and divorce. Thousand Oaks, CA: Sage.

Attorney General of Ontario. (2010). Mandatory Information Program. Toronto, ON, Canada.

Aubert

(1963). Competition and dissensus: Two types of conflict and conflict resolution. Journal of Conflict Resolution, 7, 26-42.

Barker

(2006, July 6). Homicides and suicides: National Violent Death Reporting System, United States, 2003–2004. Morbidity and Mortality Weekly Report.

Bennett

M. D.

Hughes

(2005). The art of mediation. Notre Dame, IN: National Institute for Trial Advocacy, University of Notre Dame.

Black

(2004). Violent structures. In Zahn

Brownstein

Jackson

(Eds.), Violence: From theory to research (pp. 145-158). Cincinnati, OH: Anderson.

Block

C. R.

(2000). The Chicago Women’s Health Risk Study: Risk of serious injury or death in intimate violence. A collaborative research project. Washington, DC: U.S. Department of Justice.

10.

Brassiolo

(2011). Domestic violence and divorce law: When divorce threats become credible. Paper presented at the ISNE Meeting, Dublin, Ireland.

11.

Braver

O’Connell

(1998). Divorced dads: Shattering the myths. New York: Penguin.

12.

Buchanan

C. M.

Heiges

(2001). When conflict continues after the marriage ends. In Grych

J. H.

Fincham

F. D.

(Eds.), Interparental conflict and child development: Theory, research, and applications (pp. 342-363). New York: Cambridge University Press.

13.

Buss

D. M.

(2005). The murderer next door: Why the mind is designed kill. New York: Penguin.

14.

Canadian Panel on Violence Against Women. (1993). Changing the landscape: Ending violence, achieving equality (Cat. No. SW45-1/1993E). Ottawa, ON: Minister of Supply and Services.

15.

Canadian Press. (2012, June 15). Quebec judge guilty of murdering wife. Toronto Star, p. A8.

16.

Cherlin

(2009). The marriage-go-round: The state of marriage and the family in America today. New York: Alfred A. Knopf.

17.

Conner

K. R.

Duberstein

P. R.

Conwell

Seidlitz

Caine

E. D.

(2001). Psychological vulnerability to completed suicide: A review of empirical studies. Suicide and Life-Threatening Behavior, 31, 367-382.

18.

Coser

(1956). The functions of social conflict. New York: Free Press.

19.

Crawford

Gartner

(1992). Woman killing: Intimate femicide in Ontario, 1974-1990. Report prepared for the Women We Honour Action Committee, Toronto, ON, Canada.

20.

Cummings

E. M.

Davies

P. T.

(1994). Children and marital conflict: The impact of family dispute and resolution. New York: Guilford Press.

21.

Daly

Wilson

(1988). Homicide. New York: Aldine de Gruyter.

22.

Daly

Wilson

(1995). Evolutionary perspective on male sexual proprietariness. In Rubach

Weiner

(Eds.), Interpersonal violent behaviors: Social and cultural aspects (pp. 109-133). New York: Springer.

23.

Dee

T. S.

(2003). Until death do you part: The effects of unilateral divorce on spousal homicides. Economic Inquiry, 41, 163-182.

24.

Dobash

R. E.

Dobash

R. P.

Cavanagh

Medina-Ariza

(2007). Lethal and nonlethal violence against an intimate female partner: Comparing male murderers to nonlethal abusers. Violence Against Women, 13(4), 329-353.

25.

Dugan

Nagin

D. S.

Rosenfeld

(1999). Explaining the decline in intimate partner homicide. Homicide Studies, 3, 187-214.

26.

Durham Response to Woman Abuse. (2007). Task force report on court, police and legal issues. Report presented to the Ontario Women’s Directorate, Toronto, Canada.

27.

Easteal

P. W.

(1993). Killing the beloved. Canberra: Australian Institute of Criminology.

28.

Ellis

(1994). Family mediation pilot project. Report submitted to the Attorney General of Ontario, Toronto, ON, Canada.

29.

Ellis

Anderson

(2005). Conflict resolution: An introductory text. Toronto, ON, Canada: Emond Montgomery.

30.

Ellis

Sakinofsky

Stuckless

(2012). Estrangement-associated lethal domestic violence: Risk factors and prevention. Report submitted to the Department of Justice (Family Prevention Division), Ottawa, ON, Canada.

31.

Ellis

Stuckless

(1996). Negotiating and mediating marital conflicts. Thousand Oaks, CA: Sage.

32.

Ellis

Stuckless

Smith

C. E.

(2014). Marital separation and lethal domestic violence. New York: Routledge.

33.

Emery

R. E.

(2006). The truth about children and divorce. New York: Viking.

34.

Emery

R. E.

Laumann-Billings

Waldron

M. C.

Sbarra

D. H.

Dillon

(2001). Child custody mediation and litigation: Custody, contact, and coparenting 12 years after initial dispute resolution. Journal of Consulting and Clinical Psychology, 69, 323-332.

35.

Fine

(2001). Marital conflict in stepfamilies. In Grych

J. H.

Fincham

F. D.

(Eds.), Interparental conflict and child development: Theory, research, and applications (pp. 363- 383). New York: Cambridge University Press.

36.

Flynn

Swinson

White

Hunt

Roscoe

Rodway

. . . Shaw

(2009). Homicide followed by suicide: A cross-sectional study. Journal of Forensic Psychiatry & Psychology, 20, 306-321.

37.

Gillespie

Hearn

Silverman

R. A.

(1998). Suicide following homicide in Canada. Homicide Studies, 2, 46-63.

38.

Gillis

A. R.

(1996). So long as they both shall live: Marital dissolution and the decline of domestic homicide in France, 1852-1909. American Journal of Sociology, 5, 1273-1305.

39.

Gottman

J. M.

Murray

A. D.

Swanson

C. C.

Tyson

Swanson

K. R.

(2002). The mathematics of marriage: Dynamic nonlinear models. Cambridge, MA: MIT Press.

40.

Goussinsky

Yassour-Borochowitz

(2012). “I killed her, but I never laid a finger on her”: A phenomenological difference between wife-killing and wife-battering. Aggression and Violent Behavior, 17, 553-564.

41.

Hedeem

Salem

(2006). What should family lawyers know? Results of a survey of practitioners and students. Family Court Review, 44, 601-611.

42.

Jensen

(2001). Why women kill: Homicide & gender equality. London, England: Lynne Rienner.

43.

Kandiyoti

(1988). Bargaining with patriarchy. Gender & Society, 2, 274-290.

44.

Kelly

J. B.

Gigy

Hausman

(1988). Mediated and adversarial divorce: Initial findings from a longitudinal study. In Folberg

Milne

(Eds.), Divorce mediation: Theory and practice (pp. 453-474). New York: Guilford Press.

45.

Kowalski

(2005). Spousal homicides. Family violence in Canada: A statistical profile (Cat. No. 85-224). Ottawa, ON: Statistics Canada.

46.

Kraemer

H. C.

Kazdin

A. E.

Offord

D. R.

Kessler

R. C.

Jensen

P. S.

Kupfer

D. J.

(1997). Coming to terms with the terms of risk. Archives of General Psychiatry, 54, 337-343.

47.

Kurz

(1995). For richer, for poorer: Mothers confront divorce. New York, NY: Routledge.

48.

Logan

Hill

Black

M. L.

Crosby

A. E.

Karch

D. L.

Barnes

J. D.

Lubell

K. M.

(2008).Characteristics of perpetrators in homicide-followed-by-suicide incidents: National Violent Death Reporting System—17 US States, 2003-2005. American Journal of Epidemiology, 168, 1056-1064.

49.

Mahoney

(1991). Redefining the issue of separation. Michigan Law Review, 90, 1-94.

50.

Merton

(1957). Social theory and social structure. Glencoe, IL: Free Press.

51.

Milroy

C. M.

(1993). Homicide followed by suicide (dyadic death) in Yorkshire and Humberside. Medicine, Science and the Law, 33, 167-171.

52.

Mnookin

(2010). A devilish divorce. In Mnookin

(Eds.), Bargaining with the devil: When to negotiate, when to fight (pp. 209-231). New York: Simon & Schuster.

53.

Mnookin

Kornhauser

(1979). Bargaining in the shadow of the law. Yale Law Journal, 88, 952-958.

54.

Odoms

(2001). Policy responses to couple conflict and domestic violence: A framework for discussion. In Booth

Crouter

A. C.

Clements

(Eds.), Couples in conflict (pp. 227- 239). Mahwah, NJ: Lawrence Erlbaum.

55.

Pearson

Thoennes

(1984). Mediating and litigating custody disputes: A longitudinal evaluation. Denver, CO: Center for Policy Research.

56.

Pruett

Jackson

(1999). Perspectives on the divorce process: Parental perceptions of the legal system and its effect on family members. American Academy of Psychiatry and the Law, 29, 18-28.

57.

Rimelspach

R. L.

(2001). Mediating family disputes in a world with domestic violence: How to devise a safe and effective court-connected mediation program. Ohio State Journal on Dispute Resolution, 17, 95-111.

58.

Roehl

O’Sullivan

Webster

Campbell

(2005). Intimate partner risk assessment validation study: The RAVE study, practitioners summary and recommendations-validation of tools for assessing risk from violent intimate partners. Washington, DC: U.S. Department of Justice.

59.

Rogge

R. D.

Bradbury

T. N.

(1999). Till violence does us part: The differing roles of communication and aggression in predicting adverse marital outcomes. Journal of Consulting and Clinical Psychology, 67, 340-351.

60.

Sander

M. R.

Goldberg

S. B.

(1994). Fitting the forum to the fuss: A user-friendly guide to selecting an ADR procedure. Negotiation Journal, 10, 49-68.

61.

Sbarra

D. A.

Law

R. W.

Portley

R. M.

(2011). Divorce and death: A meta-analysis and research agenda for clinical, social, and health psychology. Perspectives on Psychological Science, 6, 454-474.

62.

Scanzoni

(1982). Sexual bargaining: Power politics in the American marriage. Chicago, IL: University of Chicago Press.

63.

Schneider

A. K.

Mills

(2006). What family lawyers are really doing when they negotiate. Family Court Review, 44, 612-622.

64.

Silverman

Kennedy

(1993). Deadly deeds. Scarborough, ON, Canada: Nelson.

65.

Simmel

(1955). Conflict ( Wolff

, Trans.). Glencoe, IL: Free Press.

66.

Singh

J. P.

(2012). Problems of India’s changing family and state intervention. Patna, India: Department of Sociology, Patna University.

67.

Spanier

G. B.

Casto

(1979). Adjustment to separation and divorce: A qualitative analysis. In Levinger

Moles

(Eds.), Divorce and separation: Context, causes, and consequences (pp. 211-227). New York: Basic Books.

68.

Stevenson

Wolfers

(1999). Till death do us part: Effects of divorce laws on suicide and intimate homicide (Report funded by the MacArthur Foundation and the Social Science Research Council). Cambridge, MA: Department of Economics, Harvard University.

69.

Stone

(1993). Broken lives: Separation and divorce in England, 1660-1857. London, England: Oxford University Press.

70.

Straus

M. A.

Gelles

(Eds.). (1990). Physical violence in American families: Risk factors and adaptations to violence in 8,145 families. New Brunswick, NJ: Transaction Publishers.

71.

ver Steeg

. (2003). Yes, no, and maybe: Informed decision making about divorce mediation in the presence of domestic violence. William & Mary School of Law, 9, 145-206.

72.

Wardle

L. D.

(1994). Divorce, violence and the no-fault divorce culture. Utah Law Review, 741-775.

73.

Weiss

R. S.

(1975). Marital separation. New York: Basic Books.

74.

Wilson

Daly

(1989). Marital conflict and homicide in evolutionary perspective. In Bell

Bell

N. J.

(Eds.), Sociobiology and the social sciences (pp. 45-62). Lubbock: Texas Tech University Press.

75.

Wilson

Daly

(1992). Till death do us part. In Radford

Russell

(Eds.), Femicide: The politics of woman killing (pp. 83-98). New York: Twane.

76.

Wilson

Daly

(1993). Spousal homicide risk and estrangement. Violence and Victims, 8, 3-16.

77.

Wilson

Daly

(1994). Spousal homicide (Juristat, Cat. No. 85-002). Ottawa, ON: Centre for Justice Statistics, Statistics Canada.

78.

Wolfgang

(1958). Patterns in criminal homicide. Philadelphia: Pennsylvania Press.