Review of "Parenting and the Child's World"

There is no problem more at the heart of developmental psychology than how to explain individual differences.  Why do some children succeed in school while others do not, why are some agreeable and others not, why do some fall into criminality and others not, why do some develop stable relationships and others do not?  The list is very long.  During the first three-quarters of the twentieth century there were three dominant approaches to these issues in Anglo-American psychology, the psychoanalytic model due to Freud, the (maternal) attachment model represented by Bowlby and others, and the behaviorist approaches of Watson and Skinner.  All three shared a belief that the role of genes was not of consequence and thus that the answers were to be found in environmental conditions.  The first two agreed that, because the child was most susceptible to influence when very young, the effects of parenting outweighed all other environmental causes.  In the final quarter of the century a fourth tradition, behavioral genetics, arose to challenge the others. 

Genetic influence upon child development can be shown rather straightforwardly although the logistics of the research is formidable:  First, if genes were inconsequential, identical twins reared together would display far greater correlation for key traits than identical twins reared apart, but this is not the case.  Second, if genes were inconsequential, differences in levels for key traits between identical twins as compared to fraternal twins would not be significant, but this is not the case.  Finally, if genes were inconsequential, children adopted at birth would become more similar in interests to their adopting parents than to their biological parents, but this is not the case.  Eleanor Maccoby notes in the text under review, "Behavior geneticists have made their case … [that] … genetic factors do clearly make a significant contribution to individual differences …"(37).

But behavioral genetics had a second cherished belief to critique, that the family is the most efficacious of the environmental factors prompting individual differences.  The behavioral genetic critiques of earlier research were clearly reported in David C. Rowe's The Limits of Family Influence (Guilford Press, 1994), but it was Judith Rich Harris' The Nurture Assumption: Why Children Turn Out the Way They Do (The Free Press, 1998) that brought the matter to a head. 

The text under review contains the follow-up papers of a conference held in August 1999, a gathering planned, according to the preface, to address one of Harris'  conclusions, to wit, "the assumption that what influences children's development, apart from their genes, is the way their parents bring them up … is wrong" (xiii).  It has been a decade since Rowe's semi-popularized review of these conclusions and still Harris' statement remains shocking to read.  It is fitting then that the text would begin with chapters first by Harris and then by Rowe who together outline the case against the importance of family influence.

A common way of approaching the question of individual differences is to wonder why siblings from the same family are so unalike.  It cannot be birth order since fraternal twins are so different and it cannot be genetics alone because the personality correlation for identical twins reared together is only about 50%.  As evidence of the limited influence of family environment Harris notes first that researchers have not turned up any significant personality differences between children who have and those who do not have siblings despite the fact that the family lives of these two children are so different.  Second, no important differences have been found between children beginning daycare in early infancy and those whose first three years were with a parent.  Third, children in the U. S. raised in a non-English speaking family will adopt English and not the family language as primary.  Finally, Harris reviews the literature showing that behaviors learned and developed in one social context do not readily transfer to another.  She is critical of intervention studies that teach effective home parenting but do not objectively assess behavioral changes in other contexts such as the school.  "… my survey of the intervention literature led me to conclude that home-based interventions cannot improve children's behavior at school …" (17).

David Rowe's essay focuses on what twin and adoption research reveals about parenting efficacy.  He is critical of studies manifesting correlations between certain parenting behaviors and measures of child outcomes, e.g., that children of parents who read to them consistently manifest superior verbal skills.  These studies fail to control for genetic similarities (parent and child share genotypes that increase reading behavior), and fail to control for child-to-parent effects (the child gets read to because she is genetically disposed to love stories and so reinforces the parent who reads to her.) 

The existence and scope of child-to-parent effects turns the old assumption about the importance of the specifics of parenting styles on its head.  Lab studies show that parents pay more attention to physically attractive children (26).  Fraternal twins receive less equal parental attention than identical twins (27).  In one study, researchers found that children born of biological mothers with high risk behavior characteristics and adopted out early in life received more negative control from the adopting parents than did children of less risky mothers.  This was true despite the fact that the adopting parents had no knowledge of the risk levels (30).  The conclusion is not that the adopting parents of the high risk children were predisposed to be severe in their parenting or that parenting severity causes negative child behaviors but that the negative behaviors of the child prompted negative parenting behaviors.  Finally on the matter of the weakness of shared environment to make siblings alike Rowe notes that adopted siblings who share the same home environment from infancy have very different behaviors (31).  The conclusions suggested by the Harris and Rowe contributions are that (1) genetics plays a significant role in individual differences, (2) previous research that seemed to establish parent-to-child effects was flawed on a number of counts, and (3) there is little evidence to support long term parent-to-child effects.

In her The Nurture Assumption Harris makes important use of the work of Eleanor Maccoby, particularly her article, with John Martin, "Socialization in the Context of the Family: Parent Child Interaction" in E. M. Hetherington, ed. (1983) Handbook of Child Psychology, vol. 4. (New York: Wiley).  Harris paraphrases from the conclusions of the Maccoby and Martin literature review, "Either … parents have no effect or that they have different effects on each of their children …" (Harris 1998 p. 38).  Harris notes, somewhat dramatically to be sure, "With a stroke of the pen, Maccoby and Martin had crossed out most of the things that socialization researchers had been making a living on for decades" (39).  It seems to make sense therefore that Maccoby should follow the chapters by Harris and Rowe, though the outcome is unexpected.

She begins with the claim that textbook writer Harris' 1998 critique of the state of developmental psychology was "out of date" (35) noting that "Popular psychology books and elementary textbooks are usually somewhat behind the most recent research" (36).    This claim is hard to square with Harris' citations.  For example her critique of Baumrind's program of authoritarian, permissive, authoritative parenting styles contains references from 1989 through 1996.  Her critique of the research citing the dire effects of divorce on children contains citations mostly from the 1990's.

But Maccoby is most interested in refuting the assertion that while different (unshared) environments conspire to make children different, shared environments do not seem to affect children – or at least do not conspire to make them similar.  Part of Maccoby's argument rests upon a methodological critique of the reliance by behavioral geneticists on the concept of heritability or h².  For example the h² of IQ is .60 to .80 partly on the evidence that the IQ of a child adopted at birth will correlate with the IQ of the biological and not the adopting parent.  This leads the behavioral geneticist to conclude little or no effect of the adoptee's family life on IQ.  But Maccoby notes that the correlational nature of the h² measure hides the well-known fact that IQs of adopted children are consistently higher than those of their biological parent.  She notes further that late adopted children gain IQ points between age 4 and adolescence and that the size of the gain correlates with the socioeconomic status into which the child is adopted. (38-40).  Maccoby could also have mentioned the mysterious "Flynn effect," the fact that in some European countries IQ scores are increasing over generations at an amazing rate. (Flynn, J. R. (1987). "Massive IQ gains in 14 nations:  What IQ tests really measure".  Psychological Bulletin, 101, 171-191.  But while this argues for environmental effects upon IQ (the same arguments could be made for adult body height), it does not make a case for family causes.  Children adopted into higher socioeconomic status probably do experience a cognitively enriched home life, but they also attend better schools, have more verbal play mates, etc.  Maccoby's chapter is clear and interesting and it provides a good counter to the Harris and Rowe views. 

Sharon Landesman Ramey in chapter 4, decries the book publishers and corporations who tout new "scientifically based" ideas and products that either promise more than they can provide or raise sweeping questions about whether parents or early childhood experiences really matter.  "Such distortions serve to misguide or confuse parents …" (49).  Ramey names none of the evildoers in these regards but it would be difficult not to speculate that she has Harris' book prominently in view.  Of course no one seriously proposes that the quality of parenting does not matter to a child, except perhaps those who think that parenthood equates to shaping the enduring traits of the adult whom the child is to become.  And questioning the importance for later development of early experiences within normal ranges is hardly radical, amounting as it does to the not-very-novel idea that children are more resilient that some traditions have asserted. (for example see, Kagan, Jerome. (1998). Three Seductive Ideas. Cambridge MA: Harvard University Press.) 

But Ramey makes a point about child development research, including the search for parenting effects, that is often missed.  This is that the focus is almost entirely on the explanation of "stable traitlike qualities", intelligence or personality for example.  If one is in search of parenting effects this focus rules out many features of the adolescent that may be a function of parenting parenting practice.  She mentions the child's knowledge base, his health promoting behaviors, spiritual belief systems, art and music appreciation, and more.  The point is very important.  Two young adults could score identically on the "big five" personality sprectra of extraversion, agreeableness, conscientiousness, emotional stability and intellectual openness and still live vastly different lives if one had been enthusiastically introduced to some subset of cooking, motor repair, sailing, golf, opera, plumbing, hiking, fine dining, fishing, motorcycling, carpentry, sewing, film, religious worship, archaeology, and more, while the other had few or none of these interests. 

Philip Cowan and Carloyn Pape Cowan argue that studies of family interventions can show parent-to-child effects in ways that other approaches cannot.  This is particularly interesting because such studies are very close to an experimental model of parenting and child development research.  They note that research in the field can look at concurrent correlations such as those between authoritative parenting and academic success, or retrospective designs such as relationships between early spanking and later aggression, or prospective longitudinal studies.  But all of these have problems with the directions of causation and with controlling for genetic effects (77-81).  Intervention designs have a chance of overcoming these pitfalls and the results from the other three are useful in the framing of the intervention research .  The most interesting result from their own research is that interventions that included significant discussions of marital issues as well as parenting instruction were the most successful.  Cowan and Cowan note, "It is not only what parents do with or to the child that matters, but how the parents behave with each other." (94)

There is much more to like about the chapters in this text.  Kenneth Dodge takes on Sandra Scarr's well-known claim that individual differences in early parenting, in normal situations, have little long-term effect upon enduring traitlike individual differences. (Scarr, Sandra (1992). "Developmental Theories for the 1990s: Development and individual differences. Child Development, 63, 1-19.)  Dodge questions whether we can draw clear boundaries of normality, noting from statistics on poverty, single motherhood, prenatal care, and the lifetime sexual abuse of females (1 in 3) that Scarr's "normal" qualifier would rule out fully one half of American children.  This seems wrong on two fronts.  First his examples are faulty.  The oft-cited 1 in 3 statistic, even if true and I doubt it, refers not to children but to lifetimes, raising a child as a single parent hardly qualifies as parenting outside the ranges of normality, and the same for parenting while in poverty.  More importantly, Scarr is not referring to a statistical norm.  Her view is connected to an evolutionary viewpoint where every species must have multiple possible environments in which it can flourish in addition to the others in which it cannot.  Humans can live in good health within the Arctic Circle and at the equator.  They would hardly have survived the rigors of natural selection if as children humans required two parents, two cars and two siblings.  As Scarr puts it, "… variations among environments that support normal human development are not very important as determinants of variations in children's outcomes" (Scarr 1992 p. 4).  Dodge also criticizes Scarr's reliance upon behavioral genetic studies that themselves are tied so closely to means and variances within populations, ignoring, " … within-individual variation cross-situational variation … and between-individual dynamic processes …" (217).  His third objection is that, "… it is simply unrealistic to expect that individual differences in parenting in the first few years of life will have a unique and life-long enduring effect" (217).  My trouble with this last objection to Scarr's position is that it seems to be exactly Scarr's position.  Dodge's positive point is that what children take away from their immersion in parenting are "storylike" messages: Can adults be trusted?, Is the world safe?, Can I rely on others?  These messages are stored in memory and, in turn, guide the child's future interactions and outcomes (226).  This is an interesting shift in how to think about parenting effects, though it needs a good deal of elaboration.

There is much more in this text.  It is a collection of excellent essays that report the best of recent thinking and is excellently edited.  It would be appropriate for upper level undergraduate courses as well as an early graduate level introduction to the state of thinking about issues surrounding the parent-to-child effects of diverse child rearing practices.  The text would be interesting as well for philosophers thinking about methodological and conceptual problems surrounding the nature-nurture debates.  It certainly should be included in the collection of any good undergraduate library.

 

 

© 2004 John D. Mullen

 

John Mullen, Ph.D. is Professor of Philosophy at Dowling College, Long Island, NY.  He is author of Kierkegaard's Philosophy (New American Library, 1988).