The Nurture Assumption: Why Children Turn Out the Way They Do, Revised and Updated Paperback – February 24, 2009

This is a very difficult book to review, almost as difficult as it was to read. It is not difficult to read because of technical sophisication, but because both content and style trigger very strong emotions. Make no mistake, this is not an "unscientific" or hastily written defense of bad parenting, as has been claimed in some of the prior reviews here. Agree or not, she does handle the research with competence, discipline, and insight. She does not handle it all objectively, at least in the book. Moreover, it's questionable that the rhetorical purpose of questioning a mainstream dogma could be accomplished by a "balanced" approach to surveying the extant research.

This book challenges us to think of specific ways in which we influence our children's behavior and traits outside the home, other than through heredity. Yes, as many critics claim, we can find some. Aside from the early developmental issues which Harris acknowledges, we teach our children basic problem solving and moral reasoning that they often apply when we are not around. We also provide opportunities of various kinds for children to play, to interact with other people, to learn.

If she had recognized more of that, and written more about that, many of the harsher and more sophisticated criticisms would probably be somewhat assuaged. Her evolutionary argument about children doing what is neccessary to survive childhood is not at all inconsistent with the notion that parents do have important opportunities, resources, and survival and coping skills to provide.

Even within group socialization theory, the skills don't have to come from the group, they are selected by interaction with the group.
My bone of contention with most of the critics is that this doesn't really upend the group socialization theory. It largely shows that parental influence is less pervasive and overwhelming than the popular and social science models assume. That message would probably have inspired less virulent criticism, but also less praise from supporters and certainly would have made for less of a controversy.
Harris assuredly makes some very profound points about the foundation of the social science model of parenting. Yet it is hard to avoid the feeling that she goes way overboard in spots. The greatest challenge in reading this book is completing it with an objective enough mindset to appreciate what she is really bringing to the discussion.

Sometimes she does seem, as her many critics contend, to be, by implication, waving away parental responsibility. Yet by struggling through and giving her the benefit of the doubt, I found this criticism overturned by the end of the book. She does not outright say that parenting doesn't matter at all, she says that it matters most to the family relations, and less to traits and qualities as measured in other contexts. It matters more giving children opportunities to learn and grow in the log term and interact rather than shaping them behaviorally with praise and discipline and other classical "parenting style" points of focus.

In the process, she challenges the reader to think of ways in which we influence our children, and ways they resist that influence.
One of the most interesting points made in the book, and one often glossed over in reviews, is Harris' defense of the concept of social context as a determining factor in behavior. In other words, that we take on different roles in different situations, in much more than a trivial sense. This is a necessary and profound part of her scientific argument, though perhaps it has little impact compared to the conclusion that the effect of parenting is different than we generally assume. I fear that some profound theoretical issues like this will become victims of the more general controversy over what Harris says and implies about parenting, and some of her more extreme conclusions.

In the end, I rated this book so highly not just because it is good science writing, and because it constantly engaged me and made me think hard, but because thinking about these things and talking about them with each other is part of good parenting.

Another good read along similar lines ...
Stranger in the Nest: Do Parents Really Shape Their Child's Personality, Intelligence, or Character?

This book asks a question that is interesting to many people: what factors explain the personality differences in people?
The author starts by reporting on studies that show that about half of the differences can be explained genetically. Most social scientists would agree with that assessment.
But what about the half that cannot be explained genetically? The author deals with this in three steps.
1. The _assumption_ that parenting style matters is attacked by showing that the evidence for it is merely anecdotal. Rigorous attempts to quantify the effect of parents fail to show more than a negligible impact.
2. An alternative theory is developed. She suggests, based on evolutionary biology, that there might be a greater role for peer groups than parents in shaping personality. This is a very interesting section of the book, because even outside of the context of the theory, the observations of how groups form and interact are interesting.
3. The author tries to provide empirical support for the "group socialization" theory. Ironically, to my untrained eye, this evidence appears to be largely of the anecdotal variety derided in step 1! And nowhere is there a clear demonstration of the quantitative importance of peer groups.
I believe that the author has succeeded in raising the "group socialization" theory to the same level of plausibility as the nurture assumption. But I came away feeling that neither theory is well supported.
I suspect that we may never prove that anything other than genetic factors matter in personality. A large component of the "other half" could be measurement error. A physical characteristic, such as eye color, is a relatively well-defined concept that can be measured fairly precisely. Not so with "intelligence" or "aggressiveness." These are fuzzy concepts, measured imperfectly. The mere attempt to measure these concepts induces random variation. Imagine how difficult it would be to explain height differences if we weren't quite sure what "height" really means, and if the measurements were based on rulers with 20 percent margins of error!
Try to read the book with an open mind.