Selected Current Issues in Social Science Research Relating by RacD7s


									     Selected Current Issues in the Social Science Research Relating
                     to Post-Separation Parenting
                                         Brian J. Burke

1.     Introduction

In this paper I examine three current issues in the social science research relating to post-
separation parenting; namely:

1.     Attachment theory as it applies to overnight parenting time for young children;

2.     The approximation rule for establishing post-separation parenting time; and

3.     Joint custodial/shared parenting arrangements versus sole custodial arrangements vis-
       à-vis child development and health.

I endeavour in this paper to provide a balanced approach to reporting on the selected social
science research in connection with these three topics, and so I will discuss each of these
topics, as best I can, from "both sides" of the debate.

This paper has been drafted as an adjunct to my brief presentation at the Six Minute Family
Lawyer 2012 (which presentation can last, as the name suggests, just six minutes). The
paper is designed to provide lawyers with a relatively quick reference and starting point for
considering the issues addressed in my oral presentation. It should not be taken to be a
comprehensive analysis of the topics. It would in no way be an exaggeration to state that the
extant social science research on post-separation parenting issues is voluminous in both
range and depth, and that it is being augmented by new research all the time. The quality of
this scholarship ranges from poor to excellent. My intention is to prepare further, longer
papers for publication in which I will address, in a more comprehensive manner, the topics
treated in this short paper as well as other topics relating to social science research in
connection with post-separation parenting.
2.     Attachment Theory and Overnights for Young Children

Whether the reader of this short paper has heard of attachment theory in the context of
divorce will likely depend upon how involved he or she has been over the past several years
in what might be termed the "continuing professional education world" vis-à-vis parenting
issues. If involved in this world the reader may already have at least some basic familiarity
with attachment theory. If not involved in this world, then the reader may well never even
have heard of attachment theory, let alone considered how the theory has been applied to the
context of parenting arrangements following a separation or divorce.

While it is important to provide some background to attachment theory generally before
considering its application in the divorce context, in this short paper I can do little more than
provide a brief overview, highlighting at "lightning speed" only its broad outlines.

With that caveat in place, my "high-level" summary of attachment theory generally (i.e., not
as applied to the context of divorce) is as follows: What is now called "attachment theory"
was first formulated and developed by the British paediatrician and child psychiatrist John
Bowlby. Bowlby – drawing upon his clinic research with children as well as upon studies in
ethology – observing that the offspring of ground-dwelling primates operate in such a way
that, in order to maintain safety, they maintain physical closeness to an older and mature
adult or a few adults – hypothesized that this was associated with a highly emotional bond
between the child and the adult caregiver.1 Bowlby's colleague, the clinical and
developmental psychologist (and professor at The Johns Hopkins University) Mary
Ainsworth, developed Bowlby's theory into practice, and devised a test involving infants and
an adult caregiver, which test became known as the "Strange Situation". In the Strange
Situation, a mother and her child sit in a pleasant observational room that has toys in it, and
the child plays. The mother then leaves the room and returns on two (specifically timed)
occasions – on one occasion the child is left with another (non-related) adult, and on the

 Mary Main, Erik Hesse, and Seigfried Hesse, "Attachment Theory and Research: Overview With Suggested
Applications for Child Custody", Family Court Review, Vol. 49, No. 3, July 2011, 426-463, at p. 428-29.
second occasion the child is left alone. This causes the child to undergo stress – because it is
a certain level of stress (or a certain degree of anxiety or fear; or a certain level of pain) that
activates attachment-seeking behaviour in the infant. From the very set-up of the Strange
Situation procedure, it is not difficult to see how Dr. Ainsworth "identified what would later
become the two chief hallmarks of attachment: use of the mother as "a secure base for
exploration and play" and "flight to the mother as a haven of safety".2

In the Strange Situation, the child's behaviour – especially, but not solely, at stage of the
reunion with its adult caregiver – is very carefully observed by specially trained researchers
and that behaviour is "coded". Analyzing the Strange Situation results, Ainsworth classified
the types or styles of attachment as follows:

(1)      "Secure" – with these infants, the children engaged in play and exploration prior to
         the parental departure, showed distress upon the separation (usually crying), sought
         closeness upon reunion with the parent, and then re-engaged in play shortly after
         that.3 To be securely attached is, obviously, a good thing. According to the
         attachment researchers Hesse, Hesse, and Main:

                  These infants, termed secure (derived from the Latin, se-cure, without
                  worry, care, concern), were found to have had mothers who were
                  sensitively responsive to their signals and communications in the
                  home. Specifically, Ainsworth's narrative records showed that the
                  mothers of these infants had, for the most part (a) noted when an
                  infant signal occurred, (b) interpreted it accurately and then, (c)
                  responded promptly and, (d) appropriately.4

(2)      "Insecure" – with these infants, things did not go so well in the Strange Situation.
         Two types of insecure attachment were initially identified and categorized:

         (a) "Insecure – ambivalent/resistant": These children displayed symptoms of distress
         throughout much of the procedure and, upon reunion, their proximity-seeking

  Ibid. at 431.
  Ibid. at 432.
        involved elements of anger, and they were often distressed at such a pitch that they
        were not able to return to play upon reunion.5 According to Hesse, Hesse, and Main,
        "Ainsworth's narrative records showed that their mothers had been insensitive to their
        signals, largely through neglect, interference, and/or unpredictability".6

        (b) "Insecure – avoidant": These children acted as if they were completely detached
        from the mother – not crying or showing distress, and intentionally avoiding the
        mother upon reunion. However, this acting "as if" they were not distressed – as if
        they were, in a sense, secure – was a mask. In fact, Ainsworth and later researchers
        established that these children exhibited a quick heartbeat and high cortisol levels
        during the Strange Situation procedure, meaning that "these infants were
        behaviourally masking distress which was expressed physiologically".7

According to a 1988 meta-analysis by the attachment experts van IJzendoorn and
Kroonenberg8, the distribution of Ainsworth's three categories of attachment classifications
across various cultures in the world is as follows: "[a] majority of infants (65%) were found
to be secure, with 21% avoidant and 14% ambivalent".9

Later research added (in 1989) a fourth classification of attachment style –
"disorganized/disoriented" attachment – which can be said to map onto "odd" behaviours
exhibited by the child upon reunion with the parent in the Strange Situation; for example:

        …approaching the parent actively while simultaneously averting the head;
        falling prone upon the floor at the parent's entrance; disordered, confused and

  Ibid. at 432-33.
  Reinforced in this conclusion by the research of van IJzendoorn and Sagi-Schwartz. See IJzendoorn and Sagi-
Schwartz, "Cross-Cultural Patterns of Attachment: Universal and Contextual Dimensions", in J. Cassidy & P.R.
Shaver (eds.), Handbook of Attachment: Theory, Research and Clinical Applications (2nd ed., pp. 890-905) (New
York: The Guilford Press, 2008).
  Supra, note 1, at 433.
         indecisive movement about the room; freezing, stilling, entrance into
         temporary trance-like states.10

These odd behaviours signalling disorganized attachment were sometimes associated, it was
found, with parents who were maltreating the child. However, they were also found in
connection with parents who exhibited "not only frightening/threatening, frightened, and
dissociated behaviours in their infants' presence, but also timid-deferential, disorganized and
sexualized behaviour".11 Hesse, Hesse, and Main also note that "[i]n several studies and
meta-analyses disorganized attachment – as opposed to the avoidant and ambivalent forms of
insecure attachment identified by Ainsworth – has been found particularly predictive of later

Before leaving this extremely brief and basic exegesis on attachment theory, it is important
to point out that the concept of attachment relates to the child's reaction to a particular adult
caregiver. Put otherwise, an attachment type or style is a specific relational description, and
not a marker of the child's "temperament". In fact, at least in my view, attachment theory can
be said to be, in a sense, an "anti-temperament" index. As Hesse, Hesse, and Main state:

         Later13 – and surprisingly, to many – Strange Situation responses to mother
         and father were found to be independent. Thus, knowing an infant's
         attachment classification with one parent was entirely uninformative with
         respect to its classification with the second, so that the same infant could be
         secure with one parent and insecure with the other. This independence of
         response to the two parents was emphasized in a Berkeley study of 189
         families – which included father-infant as well as mother-infant Strange
         Situations (Main & Weston, 1981) – and reinforced the idea that Strange
         Situation responses were not indices of infant "temperament", but rather the
         product of interactions with a specific parent.14

(a)      Attachment Theory Contraindicating Overnights for Young Children Post-
         Separation: The Research of Jennifer McIntosh, Allan Schore, and Others

   Ibid. at 435-36. A fifth attachment style category, "cannot classify", is not treated in this paper.
    Ibid. at 436.
    That is, later than Mary Ainsworth's initial Strange Situation research.
    Supra, note 1, at 433.
The issue of attachment theory vis-à-vis post-separation parenting has recently come to the
fore in the family law community, perhaps principally (or, at least, most urgently) as a result
of a July 2011 special edition of the journal Family Court Review15 (the "Special Edition");
the peer-reviewed journal of the important and influential organization the Association of
Family and Conciliation Courts ("AFCC"). The Special Edition was guest edited by the
Australian psychologist and professor Jennifer McIntosh. It featured an important
monograph on attachment theory by Hesse, Hesse, and Main (parts of which are discussed
above), as well as various articles in the form of interviews between, usually, an important
attachment theorist and Dr. McIntosh.

While it would be very interesting to discuss the structure and nature of the Special Edition
in detail, as the Special Issue has became highly controversial16, that is beyond the scope of
this short paper. In this part of the paper, I will focus principally upon the Special Edition's
interview between Dr. McIntosh and Dr. Allan Schore – a prominent psychiatrist and
researcher and professor of psychiatry at UCLA, whose main work involves attachment and
neuroscience – printed as a paper in the Special Edition under the title "Family Law and the
Neuroscience of Attachment, Part I".17 Dr. Schore made a number of important statements
regarding attachment in this interview with Dr. McIntosh. I will quote some of these
statements directly and somewhat extensively so that Dr. Schore's words – not mine – are
transmitted, and also because some of Dr. Schore's statements are controversial:

1.      "In the last 15 years, two major trends of research have converged to confirm the
        basis tenets of interpersonal neurobiology. Number one, from psychology: as
        opposed to earlier models that posited the beginnings of personality between the third
        and fourth year, we now have good evidence that the prenatal and postnatal stages of
        human infancy represent the critical period for the organization of the central

   Family Court Review, Vol. 49, No. 3, July 2011.
   Especially at the AFCC's 49 th annual conference held in Chicago in the summer of 2012, at which Dr. Allan
Schore was the keynote speaker, during which he lectured on his work on neuroscience and attachment theory and
its implications on separating families.
   Allan Schore and Jennifer McIntosh, "Family Law and the Neuroscience of Attachment, Part I", Family Court
Review, Vol. 49. No. 3, July 2011, 501-12.
          dimensions of the personality. Number two, from neuroscience: the peak interval of
          attachment formation overlaps the most rapid period of massive human growth that
          takes place from the last trimester of pregnancy through the end of the 3rd year".18

2.        "We now know that the evolutionary mechanism of attachment does more than just
          provide the baby with a sense of safety and security. Rather, attachment drives brain
          development, five-sixths of which happens postnatally. In fact the brain grows more
          extensively and more rapidly in infancy that at any other stage of life. … But,
          importantly, this brain growth is influenced by "social forces" and therefore is
          "experience-dependent". It requires not only nutrients, but the emotional experiences
          embedded in the relationship it co-creates with the primary caregiver".19

3.        "There is now agreement in my field that the essential task of the 1st year of human
          life is the co-creation of a secure attachment bond of emotional communication
          between the infant and his/her primary caregiver. The baby communicates its
          burgeoning positive emotional states (e.g. joy, excitement) and negative emotional
          states (e.g. fear, anger) to the caregiver so that she can then regulate them. … This
          survival function – the capacity to communicate emotional states, and subsequently
          the capacity for self-regulation, which is the ability to regulate emotional states".20

4.        "To my mind there is one single 'primary' caregiver. A good definition of the primary
          caregiver is that, under stress, the baby moves toward this single person in order to
          seek the external regulation he/she needs at the moment. Under stress, the baby will
          usually turn to the primary caregiver, not the secondary caregivers. In most family
          settings, things are building with the father in the first year and he is definitely getting
          a good sense of who the baby is, but the primary bond in most cases is to the mother

   Ibid. at 502.
            in the first year and then, in addition to her, to the father and others in the 2nd and 3rd

5.          "In order to co-create a secure attachment, the infant seeks proximity to the primary
            caregiver, who must be subjectively perceived as predictable, consistent, and
            emotionally available. A caregiving arrangement in this essential time period that
            deprives the infant of this proximately, consistency and emotional availability will
            inhibit the development of the attachment system. Prolonged and repeated removal
            from the regulating primary caregiver not only deprives the child of an external
            coping mechanism, it also negatively impacts the ongoing maturation of the right

6.          "On the matters of a primary caregiver, neuroscience indicates that pre- and
            postnatally, the mother's right brain is the key to this role. This is time sensitive, as it
            is occurring in early critical periods. Neuroscience uses the term "critical periods",
            while development psychology uses the term "sensitive periods" (which I think
            downgrades the importance of the timing and later impact of emotional experiences
            that are needed by the right brain during the human brain growth spurt). The idea of
            50-50 custody splits in the first 2 years is to my mind, highly problematic, and will
            have negative long-term consequences. Considering the neuroscience, the data
            indicates that weekly paternal visitations in the first year would allow the father and
            baby to begin to know each other, a strong motivator for their forming an attachment.
            These contacts are important in the 2nd year, when his impact on brain development
            increases dramatically. I've written that essentially, as the mother is to the
            development of the left, because the left comes online at about a year and half, and
            generates new language functions and voluntary behaviours in a much more verbal,
            mobile toddler. I would say that if the baby is deprived of a sensitive, responsive
            father in the 2nd year, it has an even more negative effect that if that father is absent

     Ibid. at 504.
     Ibid. at 506.
        or less available in the 1st year. And it would have a more deleterious effect on the
        toddler's psychological gender formation and capacity to regulate aggression".23

7.      And in response to Dr. McIntosh's question as to whether "science can draw a line" as
        to the age at which equal parenting time makes sense, Dr. Schore replies: "I would
        say the neurobiological dividing line would be between 3 and 4 [years], the point at
        which both attachments and both hemispheres are fully functioning in the child. Even
        at this point with some children this might be too early, and so each case would have
        to be evaluated by a mental health professional".24

It is perhaps unsurprising that Dr. Schore's comments to Dr. McIntosh in this interview have
caused quite a controversy, linking as he does the primacy of mother to the child's first years
and coming at this, as he does, from the perspective of neuroscience – a so-called "hard"

A number of the other attachment theory experts interviewed in the Family Court Review
Special Edition echoed similar sentiments to Dr. Schore in terms of signalling caution, from
an attachment perspective, regarding overnights visits for young children of divorce. For
example, in response to Dr. McIntosh's comment that "[t]he 3-year-old age mark is an oft
cited cutoff, after which professionals talk about 'overnights being different'", Judith
Solomon said (in an interview by Dr. McIntosh of both Dr. Solomon and her colleague,
Carol George25):

        "[Four years of age ]…is not an arbitrary cutoff, nor is it absolute. It is a
        convenient date to point to some major achievements in development,
        including the capacity for symbolic thought an use of language which enables
        children to remember, to talk about what is going to happen, to anticipate
        changes. That is a big one. And something seems to happen in terms of self-
        regulation at that time as well. So, I think three is a convenient estimate, but a
        more precise assessment of readiness for change depends on the individual

   Ibid. at 507.
   Carol George, Judith Solomon, and Jennifer McIntosh, "Divorce in the Nursery: On Infants and Overnight Care",
Family Court Review, Vol. 49, No. 3., July 2011, 521-528.
        child. There is so much heterogeneity in development that it is very difficult
        to predict what shape a child will be in at that age".26

A similar approach – the setting out of some general principles, but with allowance made for
case-by-case analysis – was relayed by Alan Schroufe in his interview by Dr. McIntosh in
the Special Edition, which became the Special Edition article titled, "Divorce and
Attachment Relationships: The Longitudinal Journey".27 Dr. Schroufe said the following in
response to a question by Dr. McIntosh about length of separation of an infant from his or
her "primary caregiver":

        "My opinion, informed by our research is this: for babies, prior to 18 months
        of age, overnights away from the primary carer should be quite rare. It is not
        that you couldn't do something like one every two weeks. But I would not
        start with that. The situation needs to grow and change over time. So you say
        for the first 3 months, there are no overnights, but in fact, maybe even more
        frequent day contacts that we'll have later. After that first 3 months, we can do
        an overnight every few weeks, but we'll back down a little bit on the frequent
        day contact. You would get more overnights at the right pace. Why can't
        court orders be written like that? It would be very reassuring to fathers, who
        would look at this, and say, "Geez, by age 3 or 4, the child is with me a lot".
        At 3, I would not recommend it to be equal time. It is easier to see that kind of
        arrangement happening when the child is 6 or 8. I fully support the idea of
        distinctions between regular contact and equal contact, and between regular
        contact and overnight contact. It is the rhythmic, steady, reliable nature of
        contact that is needed for attachment bonds."28

(b)     The Contrary Position: Michael Lamb and Others

The McIntosh-edited Special Edition of the Family Court Review on attachment caused, as I
said, serious uproar in some segments of the family law community interested in post-
separation social science research. The debate was so intense that, one year after publishing
the Special Edition, the Family Court Review published another edition devoted in large part
to providing a forum for responses to the Special Edition29. In that edition, various other

   Ibid. at 525.
   Family Court Review, Vol. 49, No. 3., July 2011, 464-473.
   Ibid. at 472-73.
   Family Court Review, Vol. 50. No. 3. July 2012.
important social scientists voiced contrary or at least tempering views to those expressed in
the Special Edition.

Michael Lamb, a renowned researcher and psychology professor at Cambridge University30,
set out the strongest criticism of the structure of the Special Edition and also some of the
attachment views expressed in it. The abstract to Dr. Lamb's article, "A Wasted Opportunity
to Engage with the Literature on the Implications of Attachment Research for Family Court
Professionals"31, asserted that "[t]he Family Court Review Special Issue edited by McIntosh
provided a misleadingly narrow view of attachment theory and of previous attempts to
explore the implications of that theory and related research for family court professionals".32
A rather strong introduction. On the hot-button topic of there being one "primary caregiver"
for a young child, Dr. Lamb noted that the concept of "montropy" – "a term coined by John
Bowlby (1958) to describe his believe that infants form an initial or primary attachment to
their mothers or mother figures, before perhaps forming attachments to other people" – was
not founded in evidence "when Bowlby offered this speculation and…[Bowlby] changed his
belief in montropy over time".33 Noting that "over the ensuing decades, researchers have
actually studied the formation of attachments to both mothers and fathers", Dr. Lamb stated
that "[a]ll of the relevant research…suggests that infants in fact form attachments to fathers
and mothers at the same time, rather than sequentially".34 Apart from this critical point, Dr.
Lamb also pointed out that "[e]arly developmental processes are complicated, and it is
important for researchers to carefully specify whom they studied and in what circumstances,
so that they and other scholars can understand the limits to the findings' generalizability and
applicability".35 Dr. Lamb asserted that Dr. McIntosh's own 2010 study (discussed in parts
of the Special Edition) "focused only on that subset of separating families in which parents
had been unable to resolve extensive conflicts and disagreements and had resorted to

   Interestingly as well, Dr. Lamb was a former student of Mary Ainsworth.
   Supra, note 26 at pages 481-485.
   Ibid. at 482.
   Ibid. at 483.
litigation", and he claimed that Dr. McIntosh's "findings with respect to this population of
separating families with significant levels of conflict are quite consistent with the result of
other research with such parents"; but that "[c]learly these results should not be generalised
to all separating families, especially those in which the parents had shared childcare before
separation and had made postdivorce plans on their own".36 Pointing out what he perceived
as other defects in the conclusions propounded in the Special Edition, Dr. Lamb claimed that:

         Similarly, readers of the special issue would not know that the conclusions
         about damaging overnight arrangements reached in McIntosh's longitudinal
         survey study (McIntosh, Smyth & Keleher, 2010) were all based on the
         uncorroborated reports of the parents (mothers), or that similar conclusions
         about the harmful effect of overnights by Solomon and George (1999) were
         based on a study that included children whose fathers had never been involved
         in their care. The existence of other research showing that overnights proceed
         without adverse effect when children have had opportunities to establish and
         consolidate attachments to both parents before separation (Pruett, Ebling, &
         Insabella, 2004) was never mentioned in the special issue.

         Researchers and professionals recognise that we cannot talk in general about
         such issues as "the" effects of overnights without clarifying for whom, when,
         after what earlier experiences, and in which family contexts. The literature is
         replete with evidence about the importance of history in explaining the
         development and quality of children's attachment relationships, whether or not
         the parents separate … . Sadly, such qualifications, considerations, and
         cautions seldom found voice in the special issue, which portrayed a world of
         black-and-white clarity and unquestioned consensus. Because the special
         issue almost completely ignored the extensive empirical literature, it gave
         readers a completely inaccurate sense of our knowledge and understanding.37

3.       The Approximation Rule and Post-Separation Parenting Time

The second post-separation parenting theory or principle that I wish to treat in this short
paper is the approximation rule. In the American Law Institute's ("ALI") reform proposal
entitled Principles of the Law of Family Dissolution, which was released in 2002, the

   Ibid. In fairness to Dr. McIntosh it should be noted that she provided a reply to Dr. Lamb's (and others') critiques
of the Special Edition she guest edited in an article titled, "Reviewing the Opportunities: Guest Editor's Reply to
Comments" (Family Court Review, Vol. 50, No. 3, July 2012 496-501). Review of Dr. McIntosh's reply, however,
is beyond the scope of this paper.
approximation rule was put forth as the appropriate test to determine post-separation
parenting time in contested cases. The approximation rule can formulated as follows, using
the words found in paragraph 2.08(1) of the ALI's Principles of the Law of Family
Dissolution: "the court should allocate custodial responsibility so that the proportion of
custodial time the child spends with each parent approximates the proportion of time each
parent spent performing caretaking functions for the time prior to the parents' separation".
The ALI's prominence has ensured that the approximation rule has been taken very seriously
in the United States.38 As a result, a fair number of articles written by social science experts
and dealing with approximation rule have been published. In this short paper I can treat the
topic only very briefly, and so I will consider the comments of only a small number of

(a)      In Favour of the Approximation Rule: Robert Emery et al.

Perhaps the most well-known social scientist to speak out in favour of the approximation rule
is Robert Emery. Dr. Emery is a respected psychologist, a professor at the University of
Virginia, and a prolific writer and speaker in the area of post-separation parenting. In an
article titled "A Critical Assessment of Child Custody Evaluations: Limited Science and a
Flaw System", published in 2005 in the journal Psychological Science in the Public
Interest39, Dr. Emery, Randy Otto, and William O'Donohue lay heavy (and well-argued)
criticism upon the "best interests of the child" standard for judicial post-separation parenting
determinations, and advocate – among other things – the adoption of a "clear custody
standard like the approximation rule" when parents are unable to reach their own decisions
about parenting time. The authors state the following:

         In our view, the most important advantage of the approximation rule is that it
         is a clear, determinative standard. Parents and their lawyers would know what
         to expect of the courts, and this knowledge would promote settlement. In

   Notwithstanding that, to date, the only state in the United States in which the approximation rule has been adopted
is West Virginia.
   Robert Emery, Randy Otto, William O'Donoghue, "A Critical Assessment of Child Custody Evaluations: Limited
Science and a Flawed System", Psychological Science in the Public Interest", Volume 6, No. 1, 1-29.
         custody disputes that are nevertheless litigated, the approximation rule would
         sharply limit the scope of the legal inquiry, as well as any custody evaluations
         that might occur. Rather than assessing children's future best interests, under
         the approximation rule judges and custody evaluators would focus on the far
         clearer and far narrower question of each parent's past involvement in
         childrearing. …40

It must be noted, however, that the authors' endorsement of the approximation rule remains

         We also should be clear that our support of the approximation rule is
         motivated more by the problems created by the ill-defined nature of the current
         best-interests standard than by the approximation rule itself. We would be
         open to any clear and determinative rule for deciding children's best interests,
         but favour the approximation rule in its proposed over its two major rivals: (a)
         a primary caretaker parent standard, which would award sole legal and
         physical custody to the parent who did most of the childrearing; and (b) a
         presumption in favour of joint physical custody. We view the approximation
         rule as a pluralistic hybrid of these two alternatives.41

In a later paper, titled "Rule or Rorschach? Approximating Children's Best Interests"42, Dr.
Emery again qualifiedly endorses the approximation rule and states the following:

         The approximation rule is a hybrid principle. Like the primary caretaker
         standard, it looks to the past to determine future custody arrangements; like
         joint physical custody, it abandons old modes of custody and visitation and
         envisions a wide variety of parenting plans, from "traditional" every-other-
         weekend visitation to 50/50 parenting; and like the best interests standard, it
         individualizes custody arrangements for each family.

         In addition to clarity, the approximation rule offers parents a seemingly
         reasonably starting point for negotiating their own custody arrangements.43

In the recently published second edition of his book Renegotiating Family Relationships:
Divorce, Child Custody and Mediation44, Dr. Emery has re-endorsed (albeit, once again, in a
qualified manner) the use of the approximation rule for litigated cases.

    Ibid. at 22.
    Ibid. at 23.
    Child Development Perspectives, Vol. 1., No . 2, 132-134.
   Ibid. at 135.
(b)      Against the Approximation Rule: Richard Warshak, Shelley Riggs

Perhaps the most well-known critic of the approximation rule is Richard Warshak, clinical
professor of psychology at the University of Texas Southwestern Medical Center. Dr.
Warshak critiques the approximation rule on numerous grounds over the course of several
articles spanning a number of years. In his substantial 2011 article "Parenting by the Clock:
The Best-Interest-of-the-Child Standard, Judicial Discretion, and the American Law
Institute's 'Approximation Rule'"45, Dr. Warshak is cutting in his critique of the
approximation rule from the perspective of the social science research: "Steeped in
assumptions about child and family development, yet lacking reliable social science support,
the approximation rule begs for closer examination before judges and legislators embrace

Dr. Warshak's analysis of the approximation rule in this article is sophisticated and rigorous,
and it is not possible to treat his various arguments in this short paper. Focussing for the
moment on just one of his criticisms of the approximation rule, Dr. Warshak advances the
reasonable claim that the approximation rule is problematic in that it will give rise to serious
disputes about past caretaking. In this regard, he notes that tracking parenting time is
extremely difficult for social scientists – both from the perspective of sheer "time-counting",
and also in terms of ascertaining what type of work constitutes measurable and meaningful
"caretaking" or parenting-involvement, especially taking into account such extremely subtle
(but developmentally important) aspects as providing a child with a sense of security and
exhibiting empathy.47 Dr. Warshak's article is well worth reading in full.

Another critic of the approximation rule, this time from an attachment theory perspective, is
Shelley Riggs, assistant professor of psychology at the University of North Texas. In her
article "Is the Approximation Rule in the Child's Best Interests?: A Critique from the

   New York: The Guilford Press, 2012 (at pages 124-5).
   University of Baltimore Law Review (2011) Vol. 41, 83-164.
   Ibid. at 88.
   Ibid. at 128-132.
Perspective of Attachment Theory",48 Dr. Riggs first acknowledges that the approximation
rule can in one sense be seen as having some strength from an attachment theory perspective;
the principal strength being:

        …its acknowledgement of the importance of the parent-child attachment bond
        and its focus on caregiving as a critical feature of this relationship. … Thus, to
        the extent that the approximation rule encourages the continuance of
        attachment bonds and predivorce living conditions, and at the same time averts
        parental distress and conflict, it remains consistent with attachment theory

Dr. Riggs claims, though, that the approximation rule is "at odds with the attachment
literature in several crucial ways"50; namely:

1.      The approximation rule is wrong in its "assumption that the amount of time devoted
        by a parenting to caretaking responsibilities corresponds directly with parenting
        ability and the strength of emotional attachment in the parent-child relationship".51
        Dr. Riggs points out that "although time is clearly an important aspects in the
        development of attachment relationships, there are other factors that are, in fact, more
        important than the time a child spends with an attachment figure. Specifically, time
        may foster the formation of an attachment bond, but it does not necessarily foster the
        formation of a secure bond".52

2.      Where social science literature speaks of "the level of attachment or the strength of
        the attachment bond", it is in error.53 Instead, the question is the "quality or security
        of those bonds when discussing custody and visitation issues".54 As Dr. Riggs states:

             Attachment does not equal security. In other words, it is not how
             strongly attached the child is to the parent, but rather whether the

   Family Court Review, Vol. 43, No. 3, July 2005, 481-493.
   Ibid. at 486.
   Ibid. at 486-87.
   Ibid. at 487.
              child feels secure in his/her attachment to the parent that is associated
              with better child outcomes.55

Dr. Riggs concludes that "[t]hese are fundamental flaws that render the approximation rule
unreasonable and unworkable".56

4.      Joint Custody/Shared Parenting Versus Sole Custody, From a Child Development
        and Health Perspective

A very long-standing debate in the family law community is whether, from a child
development and health perspective, a joint custody/shared parenting arrangement post-
separation is superior to a sole custody/parenting arrangement. Because some of the social
science research in this area takes into account both (using the American terms) joint
physical custody (i.e., children's residential arrangements) and joint legal custody (i.e.,
decision-making), in this article I am using the term "custody" in the general and loose sense
to mean either of these. That said, it is probably the case that the principal issues being
addressed by this research are the benefits and/or negatives for children in connection with
differences in parenting time (i.e., time and/or residential arrangements), rather than
differences in decision-making power. Often, of course, as a matter of practice the former
tends to follows the latter, and perhaps vice versa.

The literature on this topic is vast and so rather than selecting just the discrete work of a
select number of researchers, in this paper I will approach the topic by considering the meta-
analytic work of one researcher and a holistic analysis conducted by another researcher.

(a)     "In Favour of" Joint Custody/Shared Parenting: Bauserman's Two Meta-

In 2002, Robert Bauserman, currently a professor of psychology at Stevenson University in
Maryland, wrote a paper in the Journal of Family Psychology titled "Child Adjustment in
Joint-Custody Versus Sole-Custody Arrangements: A Meta-Analytic Review",57 which

   Journal of Family Psychology (2002), Vol. 16. No. 1, 91-102.
became an important work. In that paper, Professor Bauserman "located[d] and meta-
analytically integrate[d] reports of child adjustment that directly compared[d] children in
joint-custody (legal and/or physical) and in sole-custody settings following divorce"58. His
meta-analysis involved 33 studies, 11 published and 22 unpublished (21 of which were
doctoral theses), spanning the period 1982 to 1999. The sample size across the studies was
1,846 sole custody children and 814 joint custody children.

Bauserman drew the following conclusions from his meta-analysis:

        Based on [the] results, children in joint custody are better adjusted, across
        multiple types of measures, than children in sole (primarily maternal) custody.
        This difference is found with both joint legal and joint physical custody and
        appears robust, remaining significant even when testing various categorical
        and continuous qualities of the research studies as moderators. … This
        finding is consistent with the hypothesis that joint custody can be beneficial to
        children in a wide range of family, emotional, behavioural, and academic
        domains. … Joint-custody children showed better adjustment in parental
        relations and spent significant amounts of time with the father, allowing more
        opportunity for authoritative parenting. The findings for joint legal custody
        samples indicate that children do not actually need to be in joint physical
        custody to show better adjustment, but it is important to note that joint legal
        custody children typically spent a substantial amount of time with the father as
        well. Importantly, a causal role for joint custody cannot be demonstrated
        because of the correlational nature of all research in this area.59

In 2012, Dr. Bauserman published a further meta-analysis in the Journal of Divorce &
Remarriage, entitled, "A Meta-analysis of Parental Satisfaction, Adjustment, and Conflict in
Joint Custody and Sole Custody Following Divorce".60 As the title of the paper suggests, in
that meta-analysis Dr. Bauserman analyzed aspects relating principally to the parents of the
underlying studies and reports. This meta-analysis involved an analysis of 50 research
reports (comprised of 30 journal articles, 14 dissertations, and 6 book chapters and books)
spanning the period 1979 to 2007.

   Ibid. at 92.
   Ibid. at 97-98.
   Journal of Divorce & Remarriage (2012), Vol. 53, 464-488.
Dr. Bauserman summarized the results of his meta-analytic findings as follows:

         In almost all areas of comparison, JC [joint custody] is associated with
         equivalent or better adjustment than MC [mother custody]. As hypothesized,
         JC fathers are more involved with their children and more satisfied with their
         relationship with the children, by both fathers' and mothers' reports; JC fathers
         are more satisfied with the child custody arrangement; JC mothers experience
         less parenting burden and stress; both JC mothers and JC fathers report less
         conflict with their ex-spouse, and more emotional support and positive
         feelings in the relationship. Frequency of relitigation is also less in cases of JC
         (especially for specific types of actions, such as child support modification).
         In contrast, JC parents and their MC counterparts did not significantly differ
         on measures of overall psychological adjustment or self-esteem, and JC
         mothers were less satisfied with the custody arrangement than were MC

(b)      "Against" Joint Custody/Shared Parenting

Many others, of course, conclude that joint custody/shared parenting is not a beneficial
arrangement for children; that it, they conclude either that it is neutral, or else that it is
negative. In this paper I will not address the criticisms of joint custody/shared parenting
made by writers such as Susan Boyd (for example, in her article "Autonomy for Mothers?
Relational Theory and Parenting Apart"62), since these deal with the topic based on a
philosophical approach as opposed to a research-based approach, and this paper deals with
matters relating to social science research.

In his article "Contact/Shared Residence and Child Well-Being: Research Evidence and its
Implications for Legal Decision-Making"63, Stephen Gilmour of Oxford University surveys
the results of certain parenting research studies conducted in various countries, principally to
answer the hypothetical question as to whether a legislative presumption of post-separation
parenting contact (i.e., not presumed joint parenting; but rather just some post-separation
contact) would be appropriate. Dr. Gilmour's conclusions from his review as to the benefits
of post-separation contact are both cautious and cautionary:

   Ibid. at 478-79.
   Feminist Legal Studies (2010) 18: 137-158.
   International Journal of Law, Policy and the Family, Vol. 20, No. 3, 344-365.
           The research reviewed in this article shows that, under correct conditions,
           contact between a child and non-resident parent can be beneficial to some
           children's adjustment. It can also be argued that such contact has potential to
           preserve links with the child's extended family, providing additional sources of
           support, and possibly more protection from abuse… . It seems reasonable,
           therefore, that law should play a role in protecting and supporting such
           relationships. It has also been suggested that law could play an educative or
           hortatory role in creating an expectation of responsible on-going

           The research evidence examined in this articles does not support adoption of a
           presumption of contact in court decision-making. Overall, the research
           suggests that it is not contact per se but the nature and quality of contact that
           are important to children's adjustment. The overall picture is of a complex
           interaction of family dynamics and demographic factors, in varying
           combinations and degrees of intensity, impacting on children's adjustment
           following parental separation. … Given that inter-parental conflict is
           potentially negatively associated with children's adjustment, the adoption of a
           presumption in disputed contact cases before the courts, with their profile of
           high conflict demonstrated in recent research…seems particularly
           contraindicated. In addition, the benefits of contact are generally modest, and
           other factors (in particular the child's resident parent) play an important role in
           moderating child well-being. As Hunt and Roberts argue, therefore, "care
           needs to be taken not to over-estimate the presumed benefits of contact"… .64

Dr. Gilmore also speaks of the importance of judicial awareness of social science research

           The research evidence suggests, therefore, that each case requires a careful
           assessment of the particular factors impacting upon it, preferably assessed with
           judicial aware of the research evidence. As Whiteside and Becker suggest:

                 Templates for decision-making need to include assessment of the
                 degree of parental cooperation, the nature and intensity of parental
                 conflict, and the quality of each parent's relationship with the child, as
                 well as the likelihood that a schedule will include enough time with
                 each parent to provide the opportunity for a meaningful relationship to
                 be sustained. (Whiteside and Becker, 2000:23)

           This last point recalls Amata and Gilbreth's findings on the benefit of
           authoritative parenting (Amata and Gilbreth, 1999). This could, of course, be

     Ibid. at 358-59.
            used as an argument for shared residence as the arrangement most likely to
            facilitate such a parental role … . The research evidence certainly does not
            support a generally negative stance towards such arrangements. Most studies
            show that child outcome is largely independent of residence arrangement, and
            some studies have shown modest benefits to child well-being of shared
            residence when compared with other arrangements. Neither, however, does
            the research support a presumption in favour of shared residence. Some
            studies have found a wide variation in child outcome within such
            arrangements; at least one study has demonstrated an increased risk in high
            conflict cases of children in shared residence feeling caught between their
            parents, with consequent negative effects on adjustment; and research showing
            that shared residence is often associated with a particular demographic profile,
            suggests that it may be suitable for some families and not others.65

Irrespective of their other merits or demerits, these conclusions expounded by Dr.
Gilmore reflect a theme found in a substantial segment of the vast post-separation
social science research regarding post-separation parenting (including studies not
treated by Dr. Gilmore in his analysis); which is that a critical indicator of healthy
adjustment for children post-separation is the absence of parental conflict.

     Ibid. at 360.

To top