Families, policy and the law

Selected essays on contemporary issues for Australia
Collection – May 2014


27. The scientists are coming: What are the courts to do with social science research?

By Michael Kearney

Download as printable PDF (463 KB)

When it comes to decisions about lives in the context of individual families, the power to make sound decisions and exercise wise judgements continues to lie in the uniquely human capacity to weigh and evaluate multiple sources of evidence. Having done that, one can only strive to reach a balanced synthesis of the facts of the matter. (Hayes, 2014, in this book)

Decisions about the parenting of children involves just such a balanced process. They rely on and are made within the context of the knowledge offered by the social sciences, both on an individual and broader basis. The manner in which social science knowledge may be used as a part of parenting determinations is a vexed and at times controversial topic.

This chapter outlines how social science knowledge is taken into account in parenting proceedings pursuant to the Family Law Act 1975 (Cth) and considers some of the issues that have troubled courts in recent times.

The underlying themes of this analysis are to:

  • acknowledge the central role that the knowledge offered by the social sciences has in determining parenting proceedings, but also recognise that such knowledge is almost always the subject of controversy, both in content and application, to any given proceedings;
  • demonstrate that the present approach to the use of such knowledge is inconsistent and, on one view, requires a degree of intellectual gymnastics if it is to be used without attracting appellate intervention; and
  • contend that the use of social science knowledge should not be confined to the prism afforded by established evidentiary principle but, rather, that there needs to be a reconsideration of the manner in which judicial officers are permitted to have regard to such material.

The role of social science research

The centrality of social science knowledge in the determination of parenting issues emerges from the provisions of Part VII of the Family Law Act itself. 1 Consideration of the statutory mandate that:

the best interests of the child are the paramount consideration

is sufficient to call to mind any number of matters that are not identifiable as propositions derived from legal principle nor the facts presented by any given case; for example, the importance (or otherwise) of bonding and attachment, and the significance of chronological and developmental age.

Such consideration also gives rise to an appreciation of the changing content of the best interests principle and the differing emphasis afforded elements of the principle over time, including in light of developing social science knowledge. The changing content of the best interests principle can be illustrated by reference to examples such as the approach of Supreme Courts to dealings with children in the parens patriae jurisdiction in the 1900s and to the so-called "mother principle" laid to rest by the Full Court of the Family Court of Australia in Raby (1976) FLC 90-104. Indeed, Kennedy (1991) went as far as to describe the best interests principle as being no more than "a somewhat crude conclusion of social policy" (pp. 90-91).

Whether to be properly understood as a legal principle or social aspiration, the application of the best interests principle in parenting proceedings is one in which social science knowledge has an integral role. Broadly, there are four overlapping means by which social science knowledge is presently being used by the courts:

  • in the application of the provisions contained in Part VII of the Act, reflecting in some part the social science knowledge which the legislature has chosen to incorporate;
  • through the evidence of experts who prepare reports addressing particular issues, usually family consultants, psychologists or psychiatrists;
  • by relying on published research and articles; and,
  • by recourse to social science research and knowledge as background or perhaps contextual material, often described as being relied upon for the purpose of understanding the material otherwise before a court.

It is the latter three considerations that inform the first in any particular case, and it is the latter two which will both underpin and inform an understanding of any expert evidence in the proceedings.

The evidence of a particular expert appointed in proceedings is available to be explored and tested in the course of a hearing in the traditional manner and by reference to such material as informed and influenced the expert. That process will not be explored further here, although it is in itself an important topic.

The present focus is the use by the court of social science knowledge that has not been prepared for or directed to an assessment of that which is in the interests of the particular child who is the subject of the proceedings.

The relevant legislative provisions

It is a fundamental aspect of parenting proceedings that issues are to be determined by the application of relevant principle to the evidence before the court. The evidence before the court, including any to be drawn from social science knowledge, is governed by the Family Law Act and the Evidence Act 1995 (Cth).

In general terms, social science knowledge will be sought to be put before a court in either the form of opinion evidence or, on occasion, pursuant to s 144 of the Evidence Act. Section 144 determines those matters of which a court may take "judicial notice" without a requirement for formal proof - for present purposes, matters of "common knowledge" - and will rarely permit social science research to be adduced (see, for example, McGregor [2012] FamCAFC 69).

Division 12A of the Family Law Act excludes the operation of parts of the Evidence Act in "child-related proceedings", incorporating the intention that parenting proceedings be determined on a less adversarial basis. As permissive as Division 12A is intended to be, however, it is important to recognise that it does not alter the fundamental proposition that proceedings pursuant to Part VII are to be determined on the basis of evidence put before the court. Thus, the Evidence Act continues to provide the "rules" or "filter" to be applied in determining the evidence that can properly be put before a court for the purpose of a parenting determination, ameliorated only to the extent provided by Division 12A.

To be admissible in parenting proceedings, social science research must be both:

  • an opinion of a specialised nature, based upon training, study or experience (s 79(1)); and
  • relevant, in that, if accepted, it "could rationally affect (directly or indirectly) the assessment of the probability of the existence of a fact in issue in the proceeding" (s 55);

It is suggested that s 69ZT(1) of Division 12A does not change the above position (see Dasreef Pty Ltd v Hawchar (2011) 243 CLR 588, [37]).

The problem to be identified here is that the present evidentiary framework does not readily permit reliance on social science research in parenting determinations, yet judicial officers are routinely having regard to that material, both formally and informally and directly and indirectly, in formulating decisions as to what is in a child's best interests. This can result in uncertain and conflicting decisions.

How the courts are using social science knowledge

One of the earliest decisions in relation to the use of social science knowledge was that of the Full Court of the Family Court of Australia in Patsalou (1995) FLC 92-580. The court considered an appeal where the trial judge's reasons for judgment had referred to a body of research relating to domestic violence, without that research being in evidence and without an opportunity being afforded to either party to make submissions or otherwise be heard in respect of it. The Full Court dismissed an appeal directed to the trial judge's use of the material on the basis that, while appearing in the reasons, there had been no reliance on the material in making any finding in the case.

Following particularly the publication of research in 2007 and 2008 (see, for example, McIntosh & Chisholm, 2007; Tucker, 2008), there was a marked resurgence in explicit reference to and reliance upon such material in judgments, particularly those in relation to interim parenting disputes. As a result, or at least at the same time, the manner in which social science knowledge was to be properly used was the subject of further attention by the courts.

Perhaps a high point was reached in 2007 in the judgment of Carmody J in Murphy [2007] FamCA 795, particularly at paras 338-342. In that case, His Honour considered the manner in which "social facts", being "facts concerning human behaviour sourced from non-legal disciplines and not specific facts to a particular case" (para. 338), could be considered by a court. Following an exhaustive review of literature and judgments on the topic, His Honour concluded that "judges are relatively free to consult accredited writings and make their own extrinsic enquiries from non-legal materials in forming and applying their own views on social issues. There is probably no other convenient or expedient way of doing it in this jurisdiction" (para. 341). Since that time, a series of decisions have sought to clarify and confine the approach to the use of social science research.

In Wheldon & Dinh [2010] FamCA 740, Murphy J engaged directly with a number of the propositions in the Murphy decision, urging caution in relation to the use of such research and concluding that published research sought to support findings in a particular case ought in the usual course be the subject of an appropriately qualified expert. Importantly, His Honour distinguished the use of such material "to support a judicial view" as being permissible. As will be developed, it is queried whether one is capable of being divorced from the other.

The Full Court has also expressed views in a series of cases as to how social science material is to be used in compliance with the tenets of procedural fairness, and in a series of decisions it has been suggested (albeit not without caution) that peer-reviewed research could be used by a decision-maker to make findings of fact if parties to the proceedings are afforded notice of an intention to do so and an opportunity to address such material (see McCall & Clark (2009) FLC 93-405; Barclay and Orton [2009] FamCAFC 159; Allen and Green [2010] FamCAFC 14). Although against this suggestion has been a series of decisions, including that in Maluka [2011] FamCAFC 72, in which the Full Court was faced with a case in which it was clear that the trial judge had expressly drawn the parties' attention to three particular articles and had provided the parties with the opportunity to make submissions in relation to that material.While the Full Court recognised that the trial judge had given notice of his intention to use the material, no notice had been given of how the material might be and was ultimately used, such that the appeal was allowed.

An evident tension has emerged in the courts as to the extent to which social science research is permitted to be taken into account and the purpose for which it may be applied, and it is suggested that the present approach of the courts is inconsistent with the realities of decision-making. Is it possible for social science research to be taken into account in making a parenting decision without relying upon it to determine issues in those proceedings? Is it possible to confine the use of social science research in proceedings to background material or the derivation of "social facts", even if such facts are distinguishable in definition and consequence from any other fact to be determined in proceedings?

These difficulties are conveniently illustrated by two separate cases (Salvati & Donato [2010] FamCAFC 263, and SCVG & KLD [2011] FamCAFC 100) in which the same Federal Magistrate, without prior notice to the parties, referred to similar social science articles in each judgment, both judgments containing the following paragraph:

This research is background material to my judgment. It is not evidence. It is not material in respect of which I take judicial notice, and I make no findings of fact as a result of this material. It is background material, and it assists in understanding the expert evidence provided by the Family Consultant. One also lives in hope that parents might learn from it. (Salvati & Donato [103]; SCVG & KLD [47])

In the first decision, of Salvati & Donato [2010] FamCAFC 263, the Full Court considered that notwithstanding the above paragraph, the magistrate had in fact impermissibly relied upon such material by considering that an order being made was contrary to the research and recommendations contained in the identified articles. In the second decision, SCVG & KLD [2011] FamCAFC 100, the Full Court found that there had in fact been no reliance upon the material in determining the proceedings and dismissed the appeal, although the court expressed the view that it would have been preferable for material not raised with the parties to not have been referred to. Notwithstanding this, the court recognised, perhaps unsurprisingly, that the "reasons for judgment would cause some disquiet in the mind of the litigant as to just how they might, consciously or unconsciously, have influenced the judicial mind" (para. 56).

How ought social science knowledge be taken into account?

The present state of the law is such that if social science research is to be relied upon either by a party or a court in determining the proceedings then it ought to:

  • form part of the evidence before the court;
  • be the subject of exploration in the oral evidence, including to the extent necessary with any expert witness; and
  • be the subject of attention in submissions, including as to the purpose to which it is to be put and any findings of fact upon which it might bear.

In the course of such process, the reliability and relevance of such material can be tested. Otherwise, a court ought not make reference to, nor rely upon, such material.

This conclusion does not, however, properly nor fully recognise the pervasive role that social science research is suggested to inevitably have in most, if not all, parenting determinations. Further, and if it be accepted that the application of the principles in Part VII to the evidence in any particular case will necessarily involve a court drawing upon such material, then the failure to identify and refer to that material gives rise to other difficulties. The more obvious of these include the source and reliability of such material and concepts of procedural fairness. These issues are all the more stark and important in interim parenting determinations where there is no ability to test contentious evidence, usually no or limited expert evidence as to the children's interests, and a greater reliance upon propositions derived from social science research as to the "best outcome" for the children.

The challenge for the legislature, courts and social scientists is to determine how social science research is to be used in the application of Part VII to the circumstances of any particular child.

The first matter to be recognised is the artificiality of seeking to draw distinctions between those matters permissibly relied upon by a court in a particular parenting determination, as opposed to those to which reference may be made for context or background or for the determination of some (but not all) facts.

This is not to seek to abandon the requirement for an evidence-based determination, but to seek to recognise that there are a multiplicity of inputs into such a decision, and to seek to confine reasons to only those matters formally in evidence before a court is to risk obscuring the real reasons for a decision in many instances.

As the Hon. Justice Michael Kirby wrote extra-judicially (1999):

The grant of power … to decision-makers who hold judicial office, ought to be conditional upon the exercise of that power in a way which the people governed by it, understand and generally accept. To keep in the dark those affected by the exercise of power and to disguise from them the true processes engaged in is the way of autocracy which fears sharing the truth with the people. My thesis is that judicial candour, although perhaps initially unsettling to those who hanker for fairytales, is more appropriate to our times. (p. 8)

The present approach to the use of social science research only serves to encourage reasons for a decision not to refer to this material at all, or to engage in the intellectual gymnastics necessary to ensure that such material as is referred to in a decision can be "quarantined" from the actual reasons for any particular decision.

The transparent disclosure of that which is to inform a judicial decision will permit not only appropriate identification of such material, but will enable engagement with questions regarding the reliability and relevance of such material by both the court and practitioners in a manner that the incorporation of such material by way of background or context (if at all) does not permit.

Perhaps the most valuable product of a clear exposition of the social science research relied upon in parenting determinations would be the facilitation of a process by which the validity and reliability of such research could be tested in a manner which conforms with existing legal aspirations for reliable and relevant evidence. The issue of whether the research is applicable or relevant to the case could also be tested.

The second matter, which arises as a consequence of the first, is the need to permit (explicit) reliance by courts upon social science research in a manner that the present state of the law does not allow. The granting of such permission would be to do no more than validate that which is already occurring. The benefit, however, of addressing the issue would be to facilitate a fuller exposition of the reasoning process, thereby permitting a critical analysis of the social science research incorporated therein.

Social science research will not often comfortably fall within the categories of "common knowledge" nor be evidence able to be efficiently and appropriately adduced in parenting proceedings in accordance with the principles of the Evidence Act. Such material does, nevertheless, form part of judicial determinations in varying measure and cannot and is not ignored. Thus, the question arising is the manner in which social science research ought to be able to be properly taken into account in parenting proceedings, not whether it should be.

One of the fundamental issues to be addressed in determining how a court is to be permitted to have regard to social science research is the manner in which the relevance and reliability of that knowledge is to be determined.

The actual bases of social science research and the methodologies by which conclusions are drawn are not often the subject of consideration in the course of parenting proceedings. Similarly, the relevance of the conclusions derived from social science research to the facts of any particular parenting case, having regard to the bases of such research, is rarely the subject of consideration. To the extent that such matters have been considered, they have been primarily the domain of expert witnesses appointed in proceedings, and the courts and practitioners have relied upon their expertise in distilling such matters.

If courts are to continue to have regard to social science research in determining parenting disputes, then it is necessary to engage in a critical analysis of the basis for the propositions emerging and to permit the parties to engage in the testing of the reliability and relevance of the research relied upon. This is not to encourage or require such an analysis to be undertaken in every case. The family law system in Australia has been so constrained in its resources that this would simply not be possible.

Courts and practitioners, no doubt informed and assisted by social scientists, need to participate in the development and understanding of a reliable and relevant body of social science knowledge that may properly be brought to bear upon parenting determinations, without the requirement for the adducing of formal evidence of such matters in particular proceedings. This is not to anticipate that such knowledge could or ought to be capable of producing binary and static propositions for ready assimilation into parenting determinations, but rather to seek to divert from a ready reliance upon recent or publicised theory solely because of such attributes. By way of further education of judicial officers and practitioners to understand and critically approach social science research, coupled with the exploration of the reliability and relevance of key research in suitable cases (whether at trial or appellate level), courts would be able to build up a body of both research and knowledge that have been both considered and found to be of value and importance in the determination of parenting disputes.

In 2000, the Australian Law Reform Commission addressed some of these issues in the report Managing Justice: A Review of the Federal Civil Justice System, and it is time to revisit the recommendations that:

  • the court processes in relation to reliance upon social science ought be made more transparent; and
  • the Family Law Act ought to be amended to make specific provision for reliance on relevant, accredited and published research in parenting proceedings.

As Altobelli FM commented in Roth [2008] FMCAfam 781, the exclusion of available social science knowledge is "inconsistent with contemporary approaches to child-focussed decision making" (para. 39). This being correct, there is a need to develop a coherent and principled approach to the use of social science knowledge in parenting determinations.


  • Australian Law Reform Commission. (2000). Managing justice: A review of the federal civil justice system (AALRC Report No. 89). Sydney: ALRC.
  • Kennedy, I. (1991). Patients, doctors and human rights. In R. Blackburn, & J. Taylor (Eds.), Human rights for the 1990s: Legal, political and ethical issues (pp. 81-108). London: Mansell.
  • Kirby, M. (1999). Judging: Reflections on the moment of decision. Australian Bar Review, 18, 4.
  • McIntosh, J., & Chisholm, R. (2007). Shared care and children's best interests in conflicted separation: A cautionary take from current research. Australian Family Lawyer, 20(1), 3-16.
  • Tucker, A. (2008). Shared parenting: Public perception vs legislative reality. Our role in making it work for children. Paper presented at the 13th National Family Law Conference, Adelaide.


1 Part VII of the Family Law Act deals with the making of orders in relation to children.