Allegations of family violence and child abuse in family law children's proceedings

A pre-reform exploratory study
Research Report No. 15 – May 2007

8. Discussion and implications

8.1 Key findings and links to prior work

8.1.1 The prevalence of allegations

At the most general level, this study supports public statements made by the FCoA over a number of years that allegations of violence and/or abuse frequently accompany post-separation child-related disputes that proceed as far as, or beyond, a formal application. Thus, in the present study, the "violence/abuse" flag (as defined in Chapter 4) was raised in the general litigants sample in 53% of the child-related applications to the FCoA and 62% of the applications to the FMC. In the judicial determination sample, these figures increased to 79% with respect to the FCoA and 67% with respect to the FMC. Broadly speaking, the findings also reflect a number of overseas reports, such as Johnston et al.'s (2005) finding that more than half the high-conflict separation cases in Californian family courts involved an allegation of spousal violence or child abuse.

When adult family violence was looked at independently142 from child abuse, the figures remained high - 49% and 57% for the FCoA and FMC general litigants sample. The figure remained the same (79%) for the FCoA judicial determination category and was almost the same (63%) for the FMC judicial determination group. When child abuse was looked at independently, this form of abuse was alleged in 23% of cases in the FCoA general litigants sub-sample and in 28% of cases in the FMC. These figures rose to 50% in the FCoA judicial determination group but, interestingly, reduced to 19% in the equivalent group within the FMC.

These findings suggest at least three processes that may be at work. First, not unexpectedly, cases that culminated in a defended final hearing had a higher percentage of violence allegations than general litigants sample cases - only a small number of which (9/225 couple cases) proceeded to full judicial determination. Second, the vast majority of child abuse allegations co-occurred with allegations of adult-to-adult violence. This trend is consistent with findings of, and suggestions by, a broad range of researchers and theorists both inside and outside the field of family law (see Chapter 1). Third, in relative terms, the data appeared to reflect a tendency to transfer family law matters containing child abuse allegations away from the FMC and into the FCoA. This may be the result of at least two factors. First, the FMC does not generally hear complex cases of more than two days' duration. Second, the FCoA has more specialised procedures for dealing with allegations of child abuse - in particular, the Magellan Program.

Overall, the data from the files examined in the Melbourne, Dandenong and Adelaide Registries supported Brown et al.'s (1998) assertion (Chapter 3) that violence allegations have become (or have become better recognised as) "core business" for the FCoA. This statement also appears to extend to the work of the FMC. As noted earlier, we cannot be certain that this finding would be closely reflected in figures from courts with family law jurisdiction in other parts of Australia.

Are the FCoA and the FMC attracting a higher percentage of violence and abuse allegations than would be anticipated in the Australian divorcing population at large? As noted in Chapter 3, Sheehan and Smyth (2000) derived data from self-reports of divorced parents in the general population. They found that, based on the legal definition of spousal violence (that is, behaviour that could be considered to be an offence under criminal law), 55% of men and 65% of women reported having experienced at least one such violent act. These figures are similar to the figures for the spousal violence and child abuse allegations in the general litigants sub-samples (FCoA: 53%; FMC: 62%) in the present study. This might suggest that the family violence allegation rates in children's proceedings in the FCoA or FMC may be similar to the reported rates of spousal violence profiles in the general divorcing population.143

At the same time, in their overview of the very extensive data derived from a series of Canadian studies, Bala, Jaffe and Crooks (2007) suggested that violence was a "factor" in about a quarter of all separations, and that cases in which violence was a factor were more likely to litigate. If these figures were extrapolated to the Australian situation, they might suggest that the critical dimension in understanding these cases is not so much whether or not violence ever occurred in the relationship (unfortunately, as noted, it would appear that violence of some sort occurs at some time in the majority of intimate partner relationships ending in divorce), but whether or not the violence is judged by one or both former partners to be an issue that impacts on the way they feel they can manage post-separation parenting arrangements.

Thus, in addition to the perceived types and frequency of violence, the way the violence has been or is being experienced, is likely to influence the decision to make an application to a court. Types and frequencies of violence can be measured, albeit imperfectly. But, as noted earlier, the way one experiences violence is highly variable and is less amenable to hard empirical analysis.

With around three quarters (79%) of adjudicated cases in the FCoA containing alleged violence, the present study's results are at the higher end of findings from other Australian studies that focused on FCoA cases.144 Thus, Rhoades et al. (2000) reported that in their sample of 78 final judgments made in the FCoA in 1998/99, 67% involved allegations of violence. In 2003, the FCoA's research into 91 child-related judgments also concluded that in 67% of cases, physical violence was an "issue".145 In just over half these cases, the Court upheld at least one of the allegations. Relying on comments made by FCoA judges in 399 contested contact disputes heard by the Adelaide Court between 1991 and 2001, Shea Hart (2004) found domestic violence to have been an issue in 46% of the cases. In 27% of these cases, a formal finding of domestic violence had been made. Finally, Kaspiew's (2005a, 2005b) qualitative analysis of 40 randomly selected child-related judgments concluded that violence was a "factor" in 58% of the cases and was "highly significant" in 25% of the judgments.

Most of the Australian studies into violence and child abuse in the context of family law are not formally comparable. Sampling techniques vary with respect to their quality and clarity. Findings are from a mixture of sources - allegations by litigants, formal determinations and less formal mentions by judges, and conclusions reached by interrogation of the documents by the researchers themselves. Different classes of individuals - litigants, judges and researchers - both define and estimate the incidence of violence. And finally, reliability processes with respect to the identification of violence are generally not described, nor are reliability figures supplied.

These limitations notwithstanding, if a process of triangulation is assumed (whereby differing methodologies and differing samples nonetheless reveal broadly similar results) and we add to this the results of the present study (which paid close attention to issues of sampling as well as to issues of validity and reliability), there is strong evidence for the proposition that allegations of violence are common in post-separation disputes over children that proceed as far as an application to one of the courts, and particularly common among the cases that continue on to the point of fully defended litigation.

8.1.2 Types of violence

As Table 5.2 reveals, the analysis of the types of spousal violence suggested that the most common allegation across all four samples related to actual physical spousal violence, followed by threats of physical violence and verbal or emotional abuse, and then property damage. Allegations of substance abuse, an established correlate of violence, were also common. Less frequent than substance abuse, but still substantial, were allegations of mental health problems.146

Prevalence of each of these allegations is consistently highest in the FCoA's judicial determination sample; indeed actual physical violence was alleged in over half the cases in this sample. The findings relating to the FCoA judicial determination sample reinforce the suggestion in the previous section that there is a general tendency to move more complex violence-related cases into the FCoA for judicial determination.147 Interestingly, the FMC general litigants sub-sample contained considerably more physical violence allegations - whether actual or threatened - and considerably more allegations of substance abuse, than its FCoA counterpart. These could be chance findings. They may, on the other hand, reflect a perception that by applying to the FMC, a more speedy result may be achieved at less cost.

Though not as common as allegations of substance abuse in any of the four sub-samples, mental health concerns were nonetheless alleged in 25% of the adjudicated cases heard by the FCoA. Mental health is a notoriously slippery area, with diagnoses being not infrequently contested by applicants and respondents, as well as by mental health professionals themselves. Certain forms of psychotic illness correlate with moderately elevated risks of violence, but to date there is little convincing evidence linking violence with other forms of mental illness.148 On the other hand, what is known in the drug and alcohol treatment field is that many clients carry a dual diagnosis of a mental illness and a drug or alcohol addiction. Which problem precedes which remains difficult to untangle.

At the level of clinical observation, as Kelly (2003) has observed, individuals with diagnoses of personality disorders, especially borderline personality disorders, not infrequently become obsessive litigators who appear incapable of understanding or empathising with the points of view of their former partners or their children. Such obsessions can be dangerous, as evidenced, for example, in Johnson's (2002) intensive study of 7 cases of familicide in the context of parenting disputes in Western Australia. Johnson's analysis led her to conclude that the 7 men in her study, all of whom had murdered their child and all but one of whom had then committed suicide, each met the American Psychiatric Association (1994) criteria for borderline personality disorder.149

As Rodgers, Smyth, and Robinson (2004) have noted, although mental health does not feature strongly in family law research and commentary, mental health issues do feature in many relationship breakdowns. Rodgers and his colleagues suggest that mental health issues are deserving of more attention in family law disputes than they have received to date. Like questions of violence and child abuse, however, the term "mental health" covers a wide range of symptoms and behaviours that, in turn, have a wide range of aetiological explanations. Clearly, the more specific the allegation, the greater the chance that its implication can be assessed during negotiations or decision-making. Clearly, too, the interaction between substance abuse and mental health issues needs careful analysis when both are alleged in parenting disputes.

8.1.3 Types of child abuse alleged

A form of parental child abuse was alleged in nearly one quarter of cases in the general litigants sample. This proportion roughly doubled in the FCoA judicial determination sub-sample, but it roughly halved in the FMC judicial determination sub-sample - a result that again suggests that the FMC is considerably less inclined to go on to hear child abuse allegations. Indeed, the FMC judicial determination sub-sample heard no cases of alleged sexual abuse, even though this category was represented in the FMC general litigants sub-sample.

Overall, the FCoA judicial determination sample's profile of allegations stood out as different, with a considerably higher proportion (nearly 40%) of the cases including allegations of actual physical child abuse and/or child sexual assault (the latter allegations were raised in 11% of cases). Interestingly, the only category in which the FMC (compared with the FCoA) judicial sample had a higher percentage of allegations was that of actual abduction. The difference was quite small (11% vs 7%), and may well reflect chance variation. Alternatively, the results may suggest a tendency to use the decision-making part of the FMC in child maltreatment allegation cases mainly when there is a perceived emergency. In such cases, there may be no time for referrals to other sources of assistance. The key need may well be a decision that is capable of containing issues as quickly as possible.150

8.1.4 Apparent severity of the alleged violence

Specifying types of violence (and their correlates) in the manner described above is only a first step towards understanding what is being alleged. For example, damage to property might be very minor, or mutual threats might be part of a couple's normal repertoire and not seen by either partner as intimidating. On the other hand, either action might be a fearful experience, clearly intended by one party to intimidate. An examination of court files with a view to coding each allegation cannot hope to capture fully the contextual circumstances in which the allegation was made. Nonetheless, it was important to assess each set of allegations on a scale of likely severity and to do this with as much transparency and inter-rater reliability as possible.

Though inevitably an imperfect exercise, the results (Table 5.4) leave us in little doubt that most of the allegations we were dealing with in this sample covered behaviours that were classified as serious (Categories B and C). Indeed, no more than one fifth of the allegations in any of the samples could be conceived as even potentially falling into a category that might be seen as less than serious violence.151

Again, too, we see a pattern with respect to the most severe allegations (Category C) that suggests a movement towards judicial determination by the FCoA. In addition, there is strong support for Brown et al.'s (1998) assertion, not just that allegations of violence are "core business" in family law disputes that go on to litigate, but that the allegations of violence are largely of a serious nature.

Two matters relating to this finding are worth noting at this point. The first is that a large minority of cases made no allegations of violence, raising important questions about the nature of the disputes in these instances and the extent to which the FCoA and the FMC are the most appropriate venues for the resolution of these disputes.152 The second is that the apparently serious nature of many of the allegations placed a significant burden on negotiators, family consultants, external experts and decision-makers, who are charged in their different ways to make sound assessments, often within short time frames. We will return to both these issues in the penultimate section of this chapter.

Finally, although the absolute numbers were relatively small, Table 5.11 suggests a gendered pattern with respect to categories of severity, with considerably more of the fathers' allegations compared to the mothers' allegations in the general litigants sample being in the least severe category (Category A).153 In addition, the only group in the judicial determination sample with a number large enough to report upon (the applicant mothers) had 71% of their allegations of violence rated in the most severe category.

8.1.5 The witnessing of family violence by children

As noted in Chapter 5, allegations of the types of violence witnessed by children followed a similar pattern to the allegations of the types of spousal violence that have already been noted. Children, therefore, were alleged to have witnessed actual physical violence most often (Table 5.6). This was followed by threatened physical violence, or emotional or verbal abuse, and then property damage. Again, these allegations were considerably more likely to have been raised in the judicial determination sub-samples and, of these sub-samples, all except emotional abuse allegations were more likely to be raised in the FCoA.

It may be useful to reflect at this point on how the child witness allegations within each sample compared with the total family violence allegations outlined in Table 5.1. As Table 5.5 indicates, virtually all cases containing allegations that a child witnessed other adult family violence, also contained allegations that the child witnessed spousal violence. Yet allegations from the general litigants sub-samples that a child had witnessed any type of family violence were made only about half as often as the allegations of family violence itself.

However, when we move into the judicial determination group, and focus on the FCoA figures in particular, there is a considerably stronger concordance between allegations of violence and the child's witnessing of that violence. Four hypotheses are worth considering with respect to this particular finding.

The first is that, because the judicial determination samples (especially the FCoA) were likely to be dealing with more complex cases, it was also likely that there would have been more opportunities for children to witness the violence as alleged. A second hypothesis relates to the fact that the FCoA judicial determination sample contained the youngest group of children. These children were more likely to be recognised by a parent as being unable to absent themselves from the violence. (They were also, as developmental psychologists would no doubt argue, the children likely to be the most negatively affected.) A third hypothesis is that one or both parents in the judicial determination group were indeed more sensitive to the impact on children of the witnessing of family violence and were therefore not willing to settle for less than a judicial hearing. A final possibility runs somewhat counter to the third. It is that, particularly in cases in which there was high acrimony, one tactic might have been to make allegations in as many domains as possible. High acrimony, especially in the context of the more adversarial processes that prevailed in 2003, is more compatible with a "winner takes all" mentality. The four hypotheses are not, of course, mutually exclusive.

Moloney and McIntosh (2006) have demonstrated how separating couples in high conflict find it difficult to remain aware of the need to protect their children from the spill-over effects of that conflict. As noted in Chapter 1, one or both parents struggle at these times to remain adequately attuned to their children's needs. This dynamic is likely to be strongly present during intermittent periods of family violence when the violence emerges primarily out of a failure to negotiate situational factors. On the other hand, an absence of parental attunement, at least on the part of the perpetrator, is likely to be chronically present when the violence is motivated primarily by an obsessive need to control one's partner (and probably one's children).154

Table 5.12 reveals that, whether a respondent or applicant, mothers in the general litigants and judicial determination samples were more likely than fathers in the same samples to allege that their child or children witnessed violence - the most common category of which was physical violence (Table 5.13). This finding may largely reflect the fact that a higher proportion of mothers than fathers alleged physical spousal abuse and therefore a higher proportion of mothers than fathers would have had the opportunity to allege that a child witnessed such abuse. It might also suggest a greater sensitivity on the part of mothers with respect to the impact on the child of witnessing violence.

A further caveat is required with respect to interpreting these data. Just as "no allegation of violence" does not guarantee its absence, "no allegation of the witnessing of violence by a child" does not mean that we are entitled to "rest easy" on that account in cases in which spousal violence is judged to have been likely. There is evidence, for example (such as in Tritt & Pryor, 2005), that many parents typically underestimate the emotional impact of negative family-related experiences on the lives of their children. This evidence supports the clinical experiences of many relationship counsellors and family therapists, who are not infrequently told by couples or individual parents that the children "don't know" about the conflict or the violence or about the fact that they are contemplating a separation. More often than not, it turns out that the children have been aware of these issues for some time but simply don't know how to express the fact or process the information (see Tonge, 1980, and McIntosh & Moloney, 2002, for examples and discussion with respect to children's drawings in this regard).

Thus, it would be naïve to accept the view that a child has not been affected when the violence has not literally been "seen" - for example, when it may be said to have taken place in another part of the house. As noted in Chapter 3, it has been increasingly recognised not only that abuse of children is strongly associated with spousal violence, but that spousal violence when children are anywhere in the vicinity runs a considerable risk of traumatising them. The witnessing of family violence has increasingly come to be recognised a form of child abuse in its own right.

8.1.6 Who alleged what?

As noted in Chapter 5 (see Table 5.7), being an applicant increased the chances that an allegation would be raised - an effect that was especially pronounced when the applicant was a father. But whether they were applicants or respondents, mothers were much more likely than fathers in the same sample to make allegations of both adult-to-adult violence and child abuse (Table 5.8). About a fifth of the allegations in the general litigants sample were mutual, but this rate was at least double in the judicial determination sample, probably suggesting a higher degree of complexity and/or intractability in these cases (Table 5.7).

As Table 5.8 demonstrates, allegations of spousal violence were more common than allegations of child abuse in each sample and in each of the 8 applicant respondent groups. Spousal violence allegations were especially pronounced in the judicial determination sample, with fathers in this group making such allegations at least twice as often as fathers in the general litigants sample. Within each sample, however, mothers alleged spousal violence more often than fathers and, within each gender, applicants alleged spousal violence more often than respondents. In addition, allegations of "other family violence" were made most often by applicant mothers in the judicial determination sample.

As noted in Chapter 1, previous research suggests that men are more likely than women to commit extremely intimidating and controlling acts of violence. In addition, Gilding (1991) has suggested that violence in families is cited by women far more often than men as a reason for leaving a marriage. It is not surprising, therefore, that allegations of spousal violence are made more often by women. What is perhaps surprising is that applicant fathers in the judicial determination sample made allegations related to spousal violence in just over half the cases.

What might this high rate of allegations by applicant fathers mean in the context of high rates of allegations by mothers? Do they reflect a history of mutual violence? Are they mainly an artefact of the high acrimony not infrequently seen in litigated cases? Are some of these fathers attempting to gain a tactical advantage by ensuring they allege violence (whether "true" or "false") in the first instance? Might it be seen as socially unacceptable for a mother to not make allegations of violence when fathers are making them (again whether or not the allegation is "true" or "false")? Again, such hypotheses are neither mutually exclusive nor exhaustive.

The types of spousal violence alleged by all groups of applicant and respondent parents largely followed the same patterns noted in Section 8.1.2 - with actual physical violence being most commonly alleged, followed by threatened physical abuse or emotional/verbal abuse, then actual property damage. In both the general litigants and judicial determination samples, each form of violence (except property damage in the judicial determination sample) was alleged considerably more often by both applicant and respondent mothers than fathers. In the judicial determination sample, applicant mothers alleged actual physical violence in 50% of the cases. Most sub-groups alleged substance abuse more often than mental health issues - a trend that was most pronounced for respondent mothers (Table 5.9).

Fathers in the general litigants sample were the least likely to raise child abuse allegations, but respondent fathers in the judicial determination sample were as likely as mothers in all four groups to allege child abuse. In addition, allegations of child abuse in the absence of allegations of family violence were rare for mother and father applicants and respondents (Table 5.6).

We have already noted that spousal violence was alleged considerably more often than children were alleged to have witnessed it. Further, the data strongly suggest that the majority of the spousal violence allegations concerned behaviour that was far from trivial. Taken together, what might these findings suggest?

It may be that some separating couples cope by partially detaching questions related to the welfare of their children from the spousal violence that they allege. It may also be that, like the courts, many parents struggle to reconcile questions of violence towards themselves, with the recognition that they do not want the other parent to be completely excluded from their children's lives, or at least do not see this as a reasonable outcome. Consistent with this hypothesis, McInnes' (2006) research (see Chapter 3) reported on the experiences of women in situations of serious violence, many of whom reported that they did not want parenting by the father to cease. Kaspiew's (2005a, 2005b) qualitative examination of court files (see Chapter 3) yielded similar findings.

In some cases, too, there might also be a recognition of some mutuality with respect to the spousal violence that precludes these applicants or respondents from "going too far" in their allegations of child abuse. Finally, some parents may be hopeful that the violence that has occurred will cease as a result of the separation and that, in this sense, the separation is likely to improve the situation, not only for themselves, but also for the children.155

All of the applicant mothers who made child abuse allegations in the judicial determination sample alleged actual physical abuse (Table 5.10). Fathers' allegations in this sample also mostly concerned actual physical abuse, but such allegations were made in a lower proportion of cases. The pattern of allegations in the general litigants sample was more varied, although, overall, allegations of any child abuse were made by a higher proportion of mothers than fathers.

Interestingly, the only allegations of sexual abuse in the judicial determination sample were made by respondent mothers. Allegations of sexual abuse were made by a small proportion of both applicant and respondent mothers in the general litigants sample. Larger samples (especially of the judicial determination sub-samples) and/or more contextual information would be needed before a realistic attempt could be made to interpret these data.

8.1.7 Evidentiary material

Table 6.1 demonstrates that in the general litigants sample not only was the proportion of cases containing allegations of child abuse considerably less than the proportion of cases containing allegations of spousal violence, but that within each case, the average number of allegations of child abuse was less than the average number of allegations made in the spousal violence cases. The average number of allegations of spousal violence made in the judicial determination cases was greater than in the general litigants sample - although most of this difference was accounted for by differences in the FCoA sub-sample. Indeed, in the FCoA judicial determination sub-sample, four or more allegations were raised in 50% of the cases. (The numbers were too small to make comparisons between the judicial determination and the general litigants groups in the case of allegations of child abuse.)

When it came to the provision of corroborative evidence, the numbers were stark. Table 6.2 shows that between 71% and 92% of the total number of spousal violence or child abuse allegations provided no corroborative evidence to support the allegations made. When corroborative evidence was supplied, it was most often (between 6% and 28% of the allegations) confined to a single piece; only 1-4% of allegations were accompanied by more than two pieces of corroborative evidence. Pieces of corroborative evidence were judged to be of relatively strong probative weight in 2-18% of the allegations.

Table 6.2 also reveals that no evidence in support of allegations was provided in 39-86% of the cases. In addition, the corroborative evidence provided was considered to be "all strong" in only 0-13% of the cases. Compared with all other sub-samples, cases in the FMC general litigants sub-sample were the least likely to make child abuse allegations, and less likely that the FCoA judicial determination to provide any strong evidence for such allegations (Table 5.1 and Table 6.2).

Compared with their male counterparts, mothers in the general litigants and judicial determination samples were more likely to provide corroborative evidence in relation to spousal violence (Table 6.3). Mothers in the combined judicial determination sample were more likely than mothers in the general litigants sample to provide corroborative evidence. The same trend applied to fathers. At the same time, the gender gap with respect to the provision of corroborative evidence was narrower for the judicial determination sample than for the general litigants sample. Interestingly, in the combined judicial determination sample, mothers were also more likely than fathers to provide corroborative evidence of relatively weak probative weight.

In Table 6.10, all sources of evidence that might support an allegation have been combined. Even when this was done, only 30-45% of allegations of spousal violence had any evidentiary support. However, of all cases that contained allegations of spousal violence, between 53% and 77% provided some form of evidentiary material for at least one of the allegations raised. Only 43% to 50% of cases reach this benchmark, however, when allegations of child abuse were considered in the general litigants sample.

From the perspective of gender and evidence from all sources (Table 6.11), roughly half the mothers in both the general litigants and judicial determination cases had at least one piece of relatively strong evidence or full admission with respect to at least one allegation of spousal violence. However, 43-44% in the general litigants sample had no corroborative evidence. Mothers were more likely to have better evidence supporting allegations of spousal violence than evidence supporting allegations of child abuse (Table 6.1.2), with 56% having no corroborative evidence regarding child abuse (compared with 44% who lacked corroborative evidence regarding their allegations of spousal violence). Compared with mothers, fathers more often made allegations of spousal violence that were supported by no evidence, but when evidence was supplied by fathers, it was highly likely to be relatively strong.

With respect to questions of severity of allegations, and its link to evidence, Table 6.13 demonstrates that the greater the apparent severity of the spousal violence allegation, the greater was the chance of having a full admission or evidence of relatively strong probative weight. But even when allegations were in the highest category of apparent severity (Category C), no evidence was provided in support of those allegations in 36% of the cases. Of the child abuse allegations, 67% were considered to be in Category C. Yet more than half of these provided no evidentiary support.

Once again, the contextual paucity of these data means that we can only speculate on their meaning. The data on which the present study was based did not enable us to identify the prevalence of long-standing coercive violence. But the title of Pizzey's (1973) seminal work, Scream quietly or the neighbours will hear, serves as a reminder that for a proportion of individuals within intimate partnerships, violence and abuse may be difficult to allege and, if alleged, evidence of that violence and abuse may be hard to come by.

Thus, those whose allegations suggest not only serious violence but also a pattern of such behaviour over time - the group likely to fit into Walker's (1984) formulation of the "battered woman syndrome" - are likely to have "suffered in silence" for too long. Obtaining corroborative evidence is likely to be very difficult when the violence has occurred over an extended period of time, potential sources of proof may be lost, witnesses (where there were any) may no longer be available, injuries may have faded and the non-physical symptoms of trauma may not be obvious.

8.1.8 Responses, denials and specificity

Table 6.5 indicates that the most common single response to the 808 allegations of spousal violence was "no response". Easily the next most common response, which occurred with respect to at least a quarter of the allegations, was full denial. The difference in the rate of denials between the general litigants and the judicial determination samples was minimal. A response to at least one allegation was more common in the FCoA judicial determination sample than in any other. But even in cases requiring judicial determination, "no response" to all allegations applied to a substantial minority.

Where responses to child abuse allegations were made, they were mostly full denials. Even here, however, "no response" clearly outweighed the full denial category in the FCoA and applied to one third of child abuse allegations in the FMC. Specifically, fathers in the general litigants sample were more likely to respond with denials to allegations of child abuse than to allegations of spousal violence (Table 6.7).156 Indeed, no father in this sample made a full or partial admission with respect to child abuse, though a small proportion (8%) did so with respect to spousal violence.

While the large number of denials is not a surprising finding, the substantial amount of overall silence as a response to most allegations is a phenomenon deserving more investigation. What cultural norms or court practices might support this absence of any response to so many allegations? At a general level, experienced negotiators know that early concessions can position one badly with respect to preliminary discussions or interim arrangements. This can be especially important in family law, where it has been shown even in early studies (such as Bordow, 1994; Horwill & Bordow, 1983) that significant alterations to orders and arrangements made at the interim stage are relatively uncommon.

Thus, legally informed processes may be an important key to understanding the phenomenon of "no response" to a large number of allegations. It may be thought, for example, that the making of too many responses, the content of which may later have to be fully or partially conceded in negotiations, is a poor or potentially dangerous approach, especially early in the process (see, for example, Cohen's (1980) research and theorising on negotiations). "No response" to the majority of allegations may leave more space for making partial concessions - especially "last-minute" concessions - that may be offered during negotiations in exchange for perceived benefits, such as more money or more time with a child.

If negotiations fail and a fully defended litigation process is likely, it might then become more important to respond to a greater number of allegations, for fear that an absence of response will be construed by a court as equivalent to an admission. This could partly explain why the percentage of "no responses", though still considerable, is appreciably less in the FCoA judicial determination sample. Finally, there may also be an expectation that if the matter proceeds very far, much will depend on the recommendation of the family report or other independent psychosocial assessment - in which case, perhaps, the fewer early concessions that might be contained in a response, the better.

Interestingly, when one looks to family reports as a possible source of independent assistance with respect to corroboration, two matters stand out. First, in the general litigants sample, the proportion of cases of alleged violence or abuse that attracted a family report was low. Second, within all family reports, partial or full corroboration of allegations happened relatively rarely and the discrediting of an allegation was almost non-existent. It would seem, therefore, that in this sample of parenting disputes in which violence or abuse had been alleged, family reports played a very minor probative role, and that their main function lay elsewhere.

Finally, the data presented in Table 6.14 demonstrate that low levels of corroboration and low levels of responses generally sat beside low levels of specificity. Indeed, only 16% of allegations in the general litigants sample of couple cases were judged to have high levels of detail. Furthermore, only 6% of high-level detailed allegations attracted high-level responses.

Perhaps not surprisingly, non-specific allegations usually elicited non-specific responses; but, in addition, about half the medium to highly specific allegations that were responded to were also answered non-specifically. Core issues around specificity of allegations of spousal violence and parental child abuse can be summarised by referring to Table 6.15, which shows that just under a third of all such allegations were non-specific and were not accompanied by, or did not elicit, any evidence; about a third had a medium level of detail and had no evidence; and about 10% had high specificity but also had no accompanying evidence. Although highly detailed allegations were uncommon (applying to 17% of all allegations), they were about twice as likely as less detailed allegations to provide at least some evidentiary material, much of which had relatively strong probative weight.

8.1.9 Violence allegations and outcomes

A data set such as this, which reveals the existence of a considerable number of worrying allegations of violence and abuse accompanied by low levels of specificity, low levels of corroborative evidence, and either denials or a complete absence of responses, inevitably poses challenges with respect to assessing outcomes. As noted in Chapter 7, when specific allegations with evidence of a high probative weight were fully litigated, the courts' orders were much more inclined to respond to the concerns raised. For example, they may have demonstrated signs of caution (such as daytime parenting arrangements only) or built in stronger protective qualifiers (such as supervised parenting). In addition, when specific allegations with evidence of a high probative weight led to consent orders, these orders also tended to be similarly responsive to the allegations.

As we have seen, however, most of the allegations in all four court sub-samples lacked supporting evidence and specificity. Where responses were made, they were mainly (and overwhelmingly in the case of child abuse allegations) in the form of full denials. In cases involving allegations that lacked evidence or specificity, the outcomes, whether fully litigated or not, were much more similar to the outcomes in the cases in which violence or abuse was not alleged.

Thus, at first glance, it would appear that if the parenting outcomes for many of the alleged violence cases were indistinguishable from the outcomes in the cases in which no violence or abuse was alleged, then both categories (alleged violence and no alleged violence) were being treated, on average, as if they were the same. One possible explanation for this is that allegations of violence were simply so ubiquitous and, on average, so difficult to assess in detail, that the impact they had on the outcomes became blunted.

There may be echoes here of issues that were raised in Chapter 1 - in particular, the calls in that chapter for greater definitional clarity with respect to family violence and child abuse. It was acknowledged in Chapter 1 that a focus on definitions that discriminate between possible types of family violence and child abuse runs the risk of being seen to tacitly endorse or trivialise what is always unacceptable behaviour. But, as noted, it has also been suggested by an increasing number of researchers (such as Johnson, 2005; Johnston et al., 2005) that it is no longer acceptable to continue to research or write in this domain as if all violence is the same.

This study cannot adequately tap into the subtleties that accompany the weighing of evidence when cases are judicially determined. Nor can it analyse the informal admissions, concessions or, perhaps, promises that might be made during the negotiations that take place with a view to resolving a matter without the need for a fully defended hearing. In other words, the data do not permit us to assess the nuances within or the nature of the determinations or negotiating processes that take place on behalf of each litigant and on behalf of their children when violence or abuse has been alleged.

In addition, we can make no claim to understand adequately the circumstances, thinking, motivation or advice that led the majority of litigants to make so many non-specific allegations which, in turn, often elicited no response. We have offered a few thoughts on these matters in the present discussion. Some, such as the difficulties experienced by those who have been in a victim role in breaking free, asserting their rights, detailing the nature of the violence or abuse and gathering evidential support are well recognised. Others, such as the idea that legal processes within a settlement-oriented family law "culture" might inhibit the making of fully fledged allegations or responses, are more speculative.157

Perhaps, too, amidst the ongoing controversies surrounding "false allegations" and "false denials" in family law disputes, practitioners and researchers may underestimate the extent to which it is no trivial matter for many litigants to make any allegations against a former partner with whom one has had children, and with whom one probably wanted to share a lifetime. The data may therefore be reflecting some ambivalence on the part of some of the litigants - perhaps a desire to set a tone that speaks of violence or abuse without being totally condemning of the individual.

A further way of approaching this issue is to consider that the fact that violence in the home has become a criminal matter does not in itself address the highly complex meaning of the behaviour. Bala et al. (2007), for example, cite a prosecutor in a Toronto domestic violence court, who describes the situation this way:

it's a crime. But you can't tell me a stranger hitting you is the same as your husband hitting you. There are just not as many factors involved. A stranger doesn't pay the mortgage; he isn't the father of your children and he's sure not someone, rightly or wrongly, that you love. (p. 22)

At this stage, the data suggest that to the extent that the information that informs litigation processes is a reflection of the information contained in the court files, much of the decision-making and negotiating in cases in which violence is alleged in Melbourne and Adelaide appears to be taking place in a climate of considerable factual uncertainty. This could mean that courts and negotiators are generally struggling to make transparent links between many of the violence-related allegations and the final outcomes. It could, on the other hand, mean that negotiations and decisions are being based largely on material that is not formally recorded on court files. If the first of these hypotheses is the case, one way forward would be to explore ways in which this material can be made clearer and more informative. If the second hypothesis has more weight, then the focus of future research would need to move further in the direction of understanding decision-making processes within the courts themselves and at the level of pre-trial negotiations conducted mainly, though not exclusively, by legal representatives.158

In either case, it is difficult to see how persisting with such a paucity of information attached to core sworn documents can be helpful. We suggest that where uncertainty predominates in a set of such core documents, its impact is most likely to be in the direction of a relative downgrading of the violence and child abuse allegations. We suggest this because if allegations of serious violence or abuse were to be reflected in the details of the parenting arrangements ordered or agreed to, one would expect these arrangements to look somewhat different to the parenting orders in the sample in which no allegations were made.

In the next section, therefore, we consider some of the ramifications of encouraging greater specificity and better corroborative evidence when allegations of violence and abuse are made. We suggest that negotiators and decision-makers (as well as separating families and their children) would benefit from helping to generate and respond to more detailed and more differentiated information. We consider some of the implications of such a change in the light of the evolving family law reforms.

8.2 Violence, child abuse and post-separation conflict: Some challenges and opportunities with respect to early differentiated screening and assessment

This research has drawn on two random samples to understand better the situation as it applied in courts with family law jurisdiction in two Australian states in 2003. Since that time, however, and following upon the publication of Every picture tells a story (Parliament of the Commonwealth of Australia, 2003), family law in Australia has embarked upon a path of significant change. In terms of dispute management processes, especially via the rollout of Family Relationship Centres, resources have shifted towards supporting community-based services that are more in touch with local needs and are charged with taking a more holistic approach to the multiple issues with which separating families present.

Under the new Division 12A of the Family Law Act, courts too are required to move towards less adversarial procedures aimed at encouraging separating parents to focus less on past misdemeanours and more on how to parent effectively into the future.

Consistent with these aspirations, the default post-separation presumption becomes one of sharing the arrangements in ways that permit children to have more than just a functional relationship with each of their parents and, in appropriate cases, more than just a functional relationship with other parental figures to whom they may be significantly attached.

At the same time, the Australian Government's Family Law Violence Strategy requires a focus on the right of children and their carers to be kept safe from harm. To this end, as noted in Chapter 1, family courts and the family law system are required to aim at "a system where allegations of family violence and child abuse are handled quickly, fairly and properly" (Attorney-General's Department, 2006a, p. 1).

There are clear tensions with respect to the twin objectives of, on the one hand, child-sensitive processes that promote substantial ongoing post-separation parenting, and, on the other hand, high-quality investigative and decision-making processes that put a premium on safety for both child and adult victims of violence. Indeed, in their report to the Canadian Department of Justice, it was suggested by Jaffe, Crooks, and Bala (2006) that contemporary family law in that country was trying to manage these very different expectations with insufficient appreciation of the fact that they were on a collision course. To avoid the collision, the authors called for nothing short of a paradigm shift.

Does the current emphasis on shared care following parental separation amount to a paradigm? And does greater sensitivity to the presence and consequences of violence and abuse amount to a paradigm shift? As noted, Bala et al. (2007), in their large-scale Canadian study, suggested that violence was a "factor" in about a quarter of separations and, not surprisingly, separated couples who nominate violence as a factor are more likely to litigate. At the same time, we have seen that, using the broadest of their definitions of "violence", Sheehan and Smyth (2000) found that in an Australian context, 65% of divorced women and 55% of divorced men reported having experienced violence in their relationship.

Assuming broad parity between Australia and Canada, it would seem that, although when asked, a majority of divorced individuals acknowledge a history of at least some violence in their relationship, most choose not to revisit this during their separation-related negotiations. This may be consistent with Jaffe et al.'s (2006) suggestion that the majority of separating couples do not seek much in the way of professional resources to make successful and responsible post-separation parenting arrangements.

As increased shared parental responsibilities reflect increased sharing of paid work time in "intact" families, various forms of shared post-separation parenting arrangements are, in turn, likely to be more frequently sought. At the same time, this study suggests that the majority of parents who go so far as making a formal application to an Australian family court allege some form of violence. In this sense, therefore, there is merit in Jaffe and his colleagues' (2006) suggestion that:

There is a critical need to move from a one-size-fits-all focus on shared parenting to a differential response focus in cases of family violence, including a comprehensive assessment by a social worker, psychologist or other mental health professional. (p. 50)

In our view, therefore, the core challenge facing the response of contemporary family law to matters involving children is this: can non-adversarial, child-sensitive dispute resolution and decision-making processes that begin with a presumption of sharing post-separation parenting arrangements work in tandem with the formal assessment and management of the risks associated with family violence?

We would support Jaffe and his colleagues (2006) in suggesting that an important step towards meeting this challenge is to begin with a more differentiated approach to questions of violence and child abuse.159 The extent to which this calls for a paradigm shift or simply the employment of more sophisticated assessment, negotiation and decision-making processes, may reduce itself to a question of semantics. More importantly, the evidence that has emerged in this study points to a need to gather and record information about alleged violence in ways that are more comprehensive, more nuanced and, where possible, better corroborated than was the case in most of those family law proceedings examined in our 2003 sample. A development in this direction would better recognise the research findings reported in Chapter 1 - that while family violence is never acceptable, not all acts of violence and abuse are the same with respect to their motivation or their impact.

As noted earlier, the data in this study suggest that issues raised in a significant number of cases in which violence and/or child abuse were alleged were being negotiated or judicially determined in an atmosphere of factual ambiguity. We have speculated briefly (but by no means comprehensively) on why this might be the case. We recognise that, in the final analysis, the dynamics that underpin such continued factual ambiguity are complex. Many factors come into play, not least of which is the fact that for many families the allegations themselves are likely to precipitate a crisis that in some cases actually increases the chances of being subjected to the very behaviour being alleged. At a personal level, too, alleging violence or abuse can require courage. At a personal level, the short-term and long-term consequences of making allegations are always uncertain

To better understand the dynamics that contribute to factual ambiguity around violence and abuse allegations in family law cases would require a more contextually based research design. At the same time, notwithstanding the gaps in our knowledge in this regard, the results of the present research strongly suggest the need for key parts of the family law system to continue to work proactively to better identify the extent and types of violence being alleged. From the perspective of safety, the core need here is the early identification of those cases that suggest a significant present or future risk to former partners and/or their children. Once identified, such cases require the speedy endorsement, either through negotiation or adjudication, of interim arrangements that, in the short term, prioritise safety over parent-child relationships. Such arrangements might, in the short term, restrict or even put a stop to the amount of time alleged perpetrators spend with their children.

When cases that might pose a significant risk are identified, it is of course critical that subsequent, more wide-ranging investigations are carried out quickly. In our view, turnaround times similar to those aimed for in the Project Magellan cases of suspected significant child abuse (roughly 2 months) should apply. If such investigations confirm that the violence or abuse is primarily systemic and controlling rather than primarily situational, and/or that it has been severe, then the safety mechanisms that were put in place in the interim will, broadly speaking, need to continue until and unless a rehabilitation plan is shown to have been effective. If subsequent investigations suggest that the violence or abuse was falsely alleged (whether deliberately or not), or was largely reciprocal or situational, and/or relatively minor in nature, the short-term interruption to the parent-child relationship, while clearly regrettable, is unlikely to cause long-term damage.

Early intervention means paying formal attention to processes of triage at key points of entry into the family law system that normally take place prior to interim determination by judicial officers. Broadly speaking, there are multiple points of early entry into the family law system - some relatively specialised and others less so. Mental health professionals, staff from women's and men's services, child support agency workers, drug and alcohol counsellors, and other counsellors and social workers from a wide range of agencies and specialist organisations, all deal with family breakdown in its early stages. So too do those associated with government-supported family law information and advice lines, and with telephone or web-based counselling services.

Ultimately, however, disputes over children associated with family breakdown require some form of negotiation between the disputants or, in the event of these negotiations being fully or partially unsuccessful, some form of adjudication. Currently, the three key points of entry that are most generally (albeit not exclusively) capable of providing professionally supported pre-adjudication negotiation facilities, combined with adequate assessment strategies, are:

  • community-based family dispute resolution service providers;
  • legally qualified family law practitioners; and
  • Family Court-based family consultants.160

We suggest that these points of entry into family law's dispute management system have both overlapping and differing functions. In the first instance, the aim of all three service types should be to make an assessment, as rapidly as possible, of their clients'/applicants' capacity to be assisted to negotiate on their own behalf and to negotiate as responsible parents, on behalf of their children. "Capacity" involves a range of issues, but the key questions that must be addressed are those that relate to violence, mental health and substance abuse.

When violence, mental health issues and substance abuse are alleged, or when these issues emerge as a result of professional supportive inquiry, clients/applicants require proactive assistance to articulate whatever can be recalled with respect to type, severity, frequency and impact and whatever can be recalled with respect to the evidence that might be available. Initial assessment of what type of violence is alleged and eliciting what evidence might be available to support the allegations is considerably enhanced by the adoption of a formally structured approach to information gathering. Indeed, in our view, the use of well-recognised, preferably validated violence assessment protocols should by now be standard practice among all professionals who see clients at an early stage in a potential family law dispute. We will consider this further a little later in this chapter.

Returning to the specific services noted above, we suggest that, following appropriate triage, professionals are then in a position to consider the particular relevance of their own service and what further actions they should take. Suggested core functions of each of the services are as follows:

  • Community-based family dispute resolution and associated services would normally assist in fully or partially resolving disputes in appropriate cases, using child-focused (see McIntosh & Moloney, 2006; Moloney & McIntosh, 2006), child-inclusive interventions (see Hewitt, 2007; McIntosh & Moloney, 2002) or related dispute resolution approaches as appropriate. They would also attend to other relevant needs (such as parent education, counselling) via facilitated community-based referrals.161 They would refer into the court system when capacity (or willingness) to negotiate responsibly is judged to be compromised, or when the issue proves to be unresolvable, even though the negotiations take place in good faith.162
  • Legally qualified family law practitioners, while advocating for their adult clients, would, when appropriate, support child-focused negotiations with the other parent or his/her representative. They might, for example, make use of collaborative law practices,163 possibly with the assistance of a dispute resolution practitioner and possibly with the assistance of a qualified child consultant. They might, on the other hand, use more traditional methods of negotiation. In cases in which they have concluded that the capacity or willingness of one or both clients to negotiate responsibly was compromised, legal representatives would generally seek urgent interim judicial determination.
  • Family Court-based family consultants would initially facilitate child-focused discussions between parties in appropriate cases that, for one reason or another, were not otherwise suitable for or not able to be dealt with by community-based agencies. They would provide child consultations as appropriate and provide or arrange for initial family assessments to judicial officers as needed.

The common emphasis in the above scenarios is, again, on early intervention. Early differentiated assessments of violence and abuse in family law cases, opens the way for more individually tailored outcomes for both parents and children. At this point, therefore, it is useful to return to the Canadian report of Jaffe et al. (2006) which, in essence, attempts to tie the lessons from the research literature into the most efficacious (or, in some cases, least destructive) practices and outcomes in each case.

A core suggestion of these authors is that as the evaluated risk to children or to a caregiver increases, presumptions of shared parenting should give way to more realistic notions of parallel parenting.164 As the risk increases further, this in turn gives way to supervised exchange, and then supervised time. At the extreme end, when the risk is judged to be unacceptable, the opportunity to continue to play a parental role may be denied completely. The report by Jaffe et al. (2006) also articulates a risk scale of family violence that draws upon and attempts to amalgamate a number of categories suggested by previous researchers such as Johnston (1994). It begins with minor and isolated acts. It then moves upward through high conflict, to violence that is more severe and frequent, and then to "battering".165 The final category, which, like the category of battering is likely to coincide with a recommended outcome of "no parental contact", is termed "terrorism/stalking" (Jaffe et al., 2006, p. 41).

Jaffe and his colleagues (2006) further suggested that outcomes need to be informed by a realistic assessment of available professional and community resources. More liberal parenting arrangements might be considered when appropriate professional interventions for victims, perpetrators and child witnesses are available. More conservative parenting arrangements need to be considered when such services are "inaccessible or inappropriate" or when there are "systemic barriers" to these services, such as poverty or language difficulties (p. 44). Finally, Jaffe et al. noted that attention must also be paid to the stage of proceedings at which allegations are made. Safety issues need to be given higher prominence during earlier negotiations or interim hearings - a stage at which "adequate information to evaluate the safety of children and adults" is more likely to be lacking (p. 47).

Jaffe et al. (2006) also provided case vignettes for each parenting category described above. They defined and described the core elements of each arrangement, and addressed the indicators and contra-indicators, as well as some of the special considerations that might apply. A "contra-indicator" for supervised access, for example, might be a situation in which it is judged that the safety offered by the supervisor may not be commensurate with the degree of assessed risk. In this case, professional supervision may be required or, if this cannot be arranged, a "no contact" order may be necessary. On the other hand, a contra-indicator for "no contact" may be a case in which the whole family system (rather than one individual) needs to change and where, despite the violence, "there is a solid foundation of a parent-child relationship" (p. 39). A "special consideration" with respect to supporting "no contact", even if violence has not been perpetrated, might be the sort of extreme case described by Johnston (2005) in which the alienation between child and parent is so extreme that forcing contact between them is simply not in the child's interests.

Consideration of the challenges and opportunities that exist in the difficult task of identifying violence and abuse and matching such identification with appropriate outcomes inevitably raises the question of resources. There is little doubt that adequate assessment in this area requires considerable skill and professionalism. Often, however, it also requires patience and time, the latter frequently being an expensive commodity.

Policymakers, it seems, must grapple with appropriate ways of responding to the following realities:

  • Many disputants who make child-related applications to family courts are in high conflict (see, for example, Jaffe, Austin, & Poisson, 1995).
  • High-conflict cases take up a disproportionately large percentage of the time of family courts. High conflict is not synonymous with violence or abuse, but many separating individuals who are in high conflict allege violence or abuse (Jaffe et al., 1995) and make conflicting allegations (Jaffe et al., 2006).
  • Extreme entrenched expressions of family violence are likely to be a gendered phenomenon. Less extreme, less entrenched expressions of family violence are less likely to be so. While national victimisation rates in Western countries look relatively similar for men and women, the differential impact of family violence on men and women means that gender remains an important consideration when decisions need to be made that take account of safety (see generally Chapter 1).
  • Parental separation has the potential to heighten or reduce the impact of family violence. When service providers intervene at times of separation, the impact will be linked to the accuracy of the assessment of the violence and the interventions that follow those assessments - especially the endorsements, recommendations and decisions about parenting arrangements.
  • While in some cases there is likely to be "outright lying" about violence, it is more often likely that "conflicting stories reflect differences in perception and understanding, and exaggeration and minimization" (Jaffe et al., 2006, p. 51).
  • Notwithstanding the above, the most carefully conducted research (such as Johnston et al., 2005) suggests that the majority of allegations of family violence are fundamentally valid. Similarly, Trocmé and Bala's (2005) extensive investigation into allegations of child abuse in separation-related parenting disputes provides strong evidence that, while false allegations are more common in family law cases than in the general population, a very large majority remain fundamentally valid.
  • The Australian evidence reported on in the previous chapters suggests that one or more individuals in more than half the cases in which a child-related dispute is brought to a family court alleged violence or abuse, much of it in the serious category.
  • Insofar as court files are capable of providing detailed knowledge of such allegations, it is clear that, although serious, most of these allegations lack specificity and detail, and also lack corroboration.

At the same time, the identification of these issues sits beside the following current understandings and important developments:166

  • While plagued with definitional issues and while not a unitary phenomenon, it is now clear that many aspects of family violence have the potential to impact seriously (though in differing ways) on children's functioning. So too does ongoing unresolved parental conflict which, when severe, is also associated with violence.
  • As noted by Jaffe et al. (2006), the availability of good quality supportive services is an important consideration in the assessment of the safety of post-separation arrangements when violence or abuse is a factor. This links to a key development within the Family Relationships Centres initiative, which has a strong emphasis on networking, and forming and maintaining associations with local supportive services.
  • The Family Court's family consultants initiative, which endorses a combination of facilitative interventions and the ability to report directly to a judicial officer, increases opportunities for rapid intervention in the early stages of alleged violence and abuse cases that make application to a court.
  • In recent times, there has been considerable development in violence risk assessment research in the areas of family (Dutton, 2006), family law (Austin, 2000) and parenting (Bancroft & Silverman, 2002), and in the development of assessment tools capable of showing "moderate accuracy in predicting violent recidivism" (Hanson, 2005, p. 214).
  • In an Australian context, Ogloff (2006a, p. 15)167 provided details of key assessment instruments with respect to areas such as spousal assault, sexual violence, adolescent violence and (importantly, in the light of the challenges of interim decision-making) short-term risk.
  • In addition, professional staff within Family Relationship Centres and other community-based family relationship services, as well as Family Court-employed family consultants, have been increasingly better supported and better trained in the use of formally developed protocols for assessing violence and abuse.168 The situation with regard to family law legal practitioners is less clear.
  • There is a growing amount of high-quality research-informed literature (Jaffe et al., 2006; Johnston, 2006) on the subject of increasingly differentiated responses to violence and abuse allegations in child-related disputes in family law.

Reflecting on the above research-informed realities and recent developments, a key question becomes: to what extent can allegations of violence and abuse in family law cases be adequately handled via an increasingly integrated use of existing resources?

Though it is well beyond the brief of the present report to provide a definitive answer to this question, it is useful nonetheless to speculate on the way in which matters might unfold as the Family Relationship Centres and other community-based services continue to expand, as the Family Court's family consultants become increasingly specialised in their new roles, and, perhaps, as judicial officers take on a less adversarial and more proactive role169 in assessing and deciding upon child-related cases brought before them.

With respect to the first development - the continued expansion of Family Relationship Centres and community-based services - it is important to recall in the present study that although individuals who approached the FCoA and FMC alleged violence or abuse in a majority of cases, no such allegations were made in a substantial minority of cases. This does not in itself rule out the possibility that violence or abuse was an issue in some of these cases. Nor, as noted above, is violence or abuse the only issue that would cause separating couples to seek a litigation pathway over community-based dispute resolution or other less formal negotiation opportunities. Furthermore, we do not know how many of these litigants had already made use of such opportunities but had failed to reach full or partial resolution.

It is fair to assume, nonetheless, that whether legally represented or not, a reasonable percentage of this potentially large number of cases would have been more expeditiously and more appropriately handled through community-based services. Logically, as both the expectations and the requirements to make use of community-based dispute resolution services continues to grow, the total number of cases using court services should be fewer. Within the court system, therefore, the percentage of those making allegations of violence and abuse would be expected to increase.170

It should be noted too that there is now clear evidence (see especially McIntosh & Long, 2006) that advances such as the child-focused and child-inclusive family dispute resolution interventions referred to earlier are capable of addressing and containing disputes in which parents are quite highly conflicted and have even at times, been situationally violent.171 Indeed, there is encouraging evidence that 12-month follow-up of child-focused and child-inclusive processes are frequently associated with decreased parental acrimony and more positive attitudes towards each other as parents.

If community-based services expand with respect to both the number and range of clients they see and if they successfully assist most of them to resolve disputes in both the shorter and longer term, courts are likely to have a smaller volume of cases to deal with. The smaller volume should enable cases to be given earlier and more intensive scrutiny. At the same time, however, the high percentage of violence and abuse allegations will place a burden on legal practitioners and court staff to find increasingly more predictable ways to discriminate between cases that pose significant risks to a parent or a child, and cases that are less likely to do so.

The pressure point for courts exercising family law jurisdiction, therefore, will increasingly become their capacity to make early and accurate assessments that can inform pre-hearing negotiations and, when required, that can assist judicial officers to make decisions that lead to responsible interim arrangements. The present research suggests that judicial officers were frequently being given inadequate levels of assistance to support them in making these very difficult decisions. As noted above, one way forward is for professionals involved in pre-court assessments and negotiations to routinely make use of high-quality, user-friendly and preferably validated violence assessment protocols that are capable of assisting in categorising and predicting violence.172 It is recognised that such protocols cannot be thought of as a substitute for clinical or legal wisdom, or clinical or legal judgment.173 On the other hand, even clinical or legal judgment of the highest quality is inevitably compromised if the information upon which it rests has not been comprehensively and systematically gathered.

At the end of the day, the extent to which litigation-related processes can be successful in both generating and responding to more detailed and, whenever possible, better corroborated evidence of family violence and abuse remains to be seen. The extent to which any barriers to achieving this will be attitudinal or skill-based or resource-driven, or a reflection - especially in the area of child abuse or ongoing debates about State and Commonwealth responsibilities - is also difficult to predict.174

What the current research confirms, however, is that the problem of family violence among prospective litigants is a significant one. What the broader contemporary research also indicates is that it is almost certainly multi-faceted. The results of this and other research emphasise the need for continuing efforts, especially at the early stages, to match each dispute with the process most likely to ensure an outcome that is fair, affordable, child-focused and safe.

8.3 Future research

Several lines of substantive and methodological inquiry warrant further investigation. Extension and replication are important foundation stones of social science research, and no single study should ever become the sole basis for policy or practice. Thus, it would be useful to replicate and extend this study with a national random sample. Courts and their registries vary in culture, demographic environs and resources, and these factors may well be related to the way that allegations and denials of family violence arise and are handled. A national study would clarify this, and enable the identification of registries that may need specialised support services. Replication of such a study after the reforms have been fully implemented would be an excellent means of evaluating the effectiveness of those aspects of the package that address violence.

Given that the present study has provided the relative proportions of allegations in each court, it would be reasonably straightforward to refine the sampling strategy, increase the sample sizes, and expand the study so that it was nationally representative of registries. Larger samples would increase the reliability of the findings, and allow comparisons to be made of some of the smaller groups that could not be examined in the present study (such as comparing the allegations raised in the judicial determination sample between the FMC and FCoA).

Second, it would be extremely useful to examine the potential links between the courts' cases and data collected by state- and territory-based child protection agencies and police. As noted earlier, state jurisdictions are the frontline for dealing with allegations of family violence and child abuse in Australia. While very basic information was noted in the present study about the involvement of such agencies in each case, the omission of state-based data (and the potential interrelationship with the Family Court and Federal Magistrates Court data) means that the scope of the study was far from comprehensive. Although the inclusion of such data represents an ambitious project, the gains are likely to be large since the triangulation of all these data sources may represent the best chance for assessing the veracity of allegations and denials.

Third, the largely "positivist" approach of counting "things" - most notably, allegations, denials, evidence and outcomes - in the current study has de-contextualised the data from human experience. There is little doubt that much suffering lies underneath the cold numbers presented in the various tables in this report (see, in particular, the Category C narratives summarised in Appendix C). There is much to be gained from speaking with the parties involved in the disputes, their legal representatives, court staff and others (such as counsellors, mediators and refuge staff) who may have information of relevance. Further Australian research into the experience of abuse that aims at improving our understanding of non-disclosure within the family law system may be of assistance.175 In addition, a project exploring the sorts of advice that lawyers give to their clients in raising and responding (and not responding) to allegations could prove very useful.

Fourth, the data in this study have been analysed largely as cross-sectional data. That is, how many cases contain x? It would be extremely useful to obtain all of the details of each case and build a narrative around each. These narratives could then be examined in a more holistic manner to build up a more contextualised analysis of the data. This is likely to require a considerable amount of time and effort, but is certain to yield important insights of the type that could only be speculated upon in the current study.

Fifth, we acknowledge that the "truth" or "falsity" of allegations (and the various nuances of veracity within this crude dichotomy) remains an issue that the present study could not address. The most extensive work to date on the subject - the various research reports of Bala and his colleagues in Canada - has concluded that in that country, false denials are considerably more common than false allegations. Nonetheless, the soundness of our judgments of the veracity or otherwise of allegations in the context of family law proceedings will continue to have an important impact on children and parents. This area remains a key question to be addressed by future research in Australia.

Sixth, this study has focused on formal allegations raised by potential litigants in a formal legal system. We do not know what percentage of separating couples who do not approach the court are in violent relationships and are unable or unwilling to seek assistance. Thus, an understanding of the "bigger picture" demands a more complex mix of methodological approaches.

Finally, we note that the Chief Justice of the Family Court, the Honourable Diana Bryant, has recently announced that the Court has started the electronic collection of additional data related to final parenting orders, including those arising from adjudicated matters (Family Court of Australia, 2006). These data include basic information about children's living arrangements, parenting time, relocation orders, family violence, child abuse, drug and alcohol issues, and mental health issues. They will allow the Court to monitor, at a broad brushstroke level, the nature and prevalence of allegations of family violence across all of its registries, examine the extent to which prevalence rates change over time, and explore any links between allegations and court outcomes. These new data are likely to play an important role in monitoring the performance of the family law reform package in relation to parenting disputes that involve courts with family law jurisdiction - especially those disputes in which allegations of family violence and child abuse are raised.

Footnotes

142 The term "independently" does not imply that adult-to-adult family violence has no negative impact on dependent children. Clearly it does.

143 Sheehan and Smyth's (2000) legal definition of "spousal violence", taken from the Australian Bureau of Statistics (1996), is: "any incident involving the occurrence, attempt or threat of either physical or sexual assault". This is a somewhat different definition than the definition of "family violence" from s60D (1) of the Family Law Act 1975 (Cth), which, until June 2006, was: "conduct, whether actual or threatened, by a person towards, or towards the property of a member of the person's family, that causes that or any other member of the person's family to fear for or to be apprehensive about his or her personal wellbeing or safety".

144 Findings from overseas studies appear to be more variable, with figures as low as 22% (Smart et al., 2003; though this study also included non-adjudicated cases) and as high as 83% (Sorenson et al., 1995).

145 It is unclear how the Family Court coded or analysed its data. It may not have had the research capacity and resources to scrutinise each page of each individual court file - the approach we were able to adopt.

146 It should be noted that allegations of substance abuse and mental health issues in this study were only recorded in cases that also contained allegations of family violence or child abuse. It is probable that substance and mental health issues were also raised in at least some of the cases that contained no allegations of violence or child abuse.

147 In the end, of course, only a small proportion is finally dealt with in this manner.

148 For a detailed analysis of this issue, see Link et al. (1992).

149 Although this was a total population drawn over a 10-year period, the number of men was only 7. In addition, because all but one of the cases were murder-suicides, the analysis of what may have caused this extreme behaviour was necessarily retrospective.

150 It should be noted that overall rates of allegations of child abduction were not derived because such allegations were only noted where they accompanied allegations of family violence or child abuse.

151 We acknowledge that without further contextual understanding, it cannot of course be assumed that allegations in Category A were not serious. On the other hand, it seems reasonable to assume that most of the allegations in Categories B and C, by their very nature and regardless of context, would have to attract a default presumption of being serious.

152 Again, the caveat here is that it cannot be assumed that no allegation equates with no violence, or on the other hand that alleged circumstances necessarily reflect reality. In addition, we are not implying here that only violence and abuse cases require judicial determination. With court and judicial resources stretched, however, the question of what allegations are being made by litigants who do not allege violence, and what other resolution options (such as referral to Family Relationship Centres) might be available, is worthy of further study.

153 This may be associated with the fact that fathers' allegations also tended to be less specific.

154 As noted in Chapter 1, high conflict can spill over into violence. Equally, however, some acts of violence have nothing to do with the resolution of conflict.

155 Again, this is not to downplay the findings noted in Chapter 1 that in some circumstances separation can be an especially dangerous period, during which, for example, opportunities for violent acts arise around parenting arrangements, especially during changeover times.

156 There were too few cases where fathers alleged that mothers had abused the children to assess mothers' responses to such allegations.

157 Though such speculation has support. See, for example, Kimm (2006), who has written on "lawyers' settlement conventions" in the context of Australian family law.

158 See previous studies in this regard, such as Ingleby (1992) and Eekelaar, Maclean and Beinart (2000).

159 Taken literally, there is a presumption in this suggestion that present processes in Australia are essentially the same as they were in 2003. Clearly, since 2003, the Australian Government has made strenuous efforts to better match service requirements to differing needs - the pre-court Family Relationship Centres and the court-based family consultants being two critical initiatives in this direction. We argue below that the call for more proactive attempts to differentiate forms of violence and abuse is highly compatible with these important initiatives.

160 Recent initiatives aimed at increasing access to family dispute resolution services via telephone or the Internet may also become an important entry point. New practice and assessment issues are being raised by the contemplation of such services. Because of their rapidly evolving nature, we have decided not to include them in the discussion that follows. However, see a recent review by Conley Tyler and McPherson (2006).

161 Also known as "warm referrals".

162 Relocation disputes can at times fall into this category.

163 See Family Law Council (2006).

164 "Parallel parenting" is a term that has been in use for a number of years (see Maccoby & Mnookin, 1992). According to Jaffe and his colleagues (2006) the concept developed in response to a recognition that some separated couples are competent parents, even though they remain in high conflict. In parallel parenting, "the arrangement is structured to minimize contact between the parents and protect the children from exposure from ongoing parental conflict, typically by having each parent make day to day decisions independently from each other when the children are in his or her care. ... Rather than encourage co-parenting, the goal of the plan is to disengage the parents from each other and their long-standing hostilities"(Jaffe et al., 2006, p. 34).

165 Drawing on Frederick and Tilley's (2001) work, Jaffe and his colleagues (2006) suggested that battering goes beyond violence and abuse to include "a larger system of intimidation, control and isolation that purposefully puts the victim at a power disadvantage" (p. 11).

166 Though important, this is not, of course, an exhaustive list.

167 Ogloff's (2006b) empirical work also extends to a vexed question in family law (see, for example, Kelly, 2003) - that of personality disorders and their possible links with the risk of violence.

168 See, for example, the screening and assessment guidelines recommended by the Attorney General's Department (2006b).

169 See the reference to less adversarial trials (LAT) on the website of the Family Court of Australia and the generally very positive evaluation by McIntosh (2006) of a pilot LAT program. However, as Hunter (2006) pointed out, there are tensions between LAT's laudable aim of providing a positive forum for former partners to consider their future roles as parents, and the obligation of the courts to formally address allegations of violence and abuse.

170 This assumes no significant changes in the rates of separation of parents with children, and no significant changes in rates of child-related disputation and violence-related allegations following separation.

171 Under the family law reforms, individuals who allege violence or abuse will not be required to use community-based dispute resolution services. It is likely, however, that for a variety of reasons, some individuals and couples who have experienced violence will opt for such services. For their part, dispute resolution practitioners will continue to have an obligation to determine the appropriateness of such a process and, on the occasions when they judge that each party has the capacity to represent themselves and their children adequately despite some evidence of past violence, to ensure that they do so safely.

172 An application to a court accompanied by a certificate of referral from a Family Relationship Centre or similar agency or individual might possibly act as an additional alert to court staff.

173 For example, good practitioners recognise that details of serious violence and abuse are more likely to be revealed in the context of a trusting relationship. This takes time. Having met this first condition, the questions about violence and abuse then need to be specific and asked at a pace that the individual can handle (remembering that if violence or abuse has occurred, the very questions can inadvertently re-traumatise).

174 This last issue requires ongoing careful analysis. We note the Family Law Council's (2002) recommendation that a Federal Government Child Protection Service be formed. At the same time, we are aware that Project Magellan, which, in part at least, addresses the problems identified by the Family Law Council, has been expanded throughout Australia and is currently being evaluated. Western Australia too has responded to its unique set of circumstances with its carefully researched Columbus project (see Murphy & Pike, 2003).

175 We recognise that much good work has already been done on questions related to reluctance to disclose violence and abuse. It may be that some of this research could be re-worked with a particular focus on allegations made as part of parenting disputes. Or it may be that further in-depth explorations of how litigants make decisions about what and how much to allege (and perhaps not allege) would be a valuable additional avenue of inquiry.