Book Review: "Selfish Reasons to Have More Kids"

I’m bemused. I’ve just been reading the anti-Amy-Chua, Bryan Caplan, whose “Selfish Reasons to Have More Kids” declares that parenting should not be hard or anxiety-provoking, because what parents do doesn’t have much long-term impact anyway. I must say it’s not so easy to make Caplan’s claims and citations congruent with much that’s known about child development. I wouldn’t want to say he’d been cherry-picking, but he seems to have used a peculiar pie recipe, and some of the fruit is hard to identify.

Let me begin my discussion of “Selfish Reasons” by saying that Caplan starts and ends with a surprising and ill-supported premise: that it’s better to have more people, better for the whole world and for you, too, even if they aren’t your children. If we don’t have lots of people, he says, much of the world will be like living in Hays, Kansas (Caplan’s example, not mine, and what he means is that it would be boring to him). This statement suggests some interesting ideas, because what if there were suddenly many fewer people? This has happened at times, for instance in the time of the Black Death in Europe, and the social consequences were enormous. But you can’t necessarily reason from one event’s results to the possible consequences of its opposite. Having no money is bad, and we can reason from that to the idea that lots of money is good. On the other hand, what about Vitamin A? A deficiency is a bad thing and can result in blindness-- a lot is also bad, as we see from the effects on Arctic explorers of eating polar bear livers.

Have we forgotten that much population growth would mean one of two things: either an increased and destructive demand for food and energy, or a great many starving and distressed people? To assert that it’s better to have more people, without providing a rationale that addresses these possibilities, is certainly to slide by an essential question.

However, I can accept with enthusiasm the idea that having more children [than you thought you’d want] would have a number of selfish benefits for the individual or couple. Yes, kids are a lot of fun, and in the words of Holly Near’s old song, “they remind us how to play”. Having a little crowd of people you know very well is a great idea in this age of high mobility, where the old gang of old friends may not be easy to reach. It’s great to see the grandchildren and watch the whole process of parenting from an intimate but safe distance.

Okay, so far we have two assertions by Caplan—first, that it’s good for the world to have more people (which I challenge), and second, it would be nice for me and others like me, and probably for my children and grandchildren too, to have a big family (which I accept).

But how about the assertion that it doesn’t really matter in the long run what parents do? Ah, here’s the crux, because now we run into matters of definition and of choices of information. Everyone is going to die, no matter what their parents did, so looking at occurrence of death would support the idea that parenting makes no difference, but naturally that’s not what Caplan or Judith Rich Harris mean when they make their claims.

I’m going to make a suggestion about what Caplan really does mean. I think he means that the small differences in childhood experiences for the families he knows don’t over-ride the effects of biological differences. Speaking about a highly restricted range of experiences, Caplan is saying that within that range experiences don’t have much long-term effect. Whether the child goes to a Montessori school or a Waldorf school doesn’t make much difference. Whether he’s given a time-out or a single smack on the behind doesn’t matter much in the long run. But Caplan knows that there are some things that do make a difference, because he points out (on p. 89 and elsewhere) that you ought to be kind to your children. He thinks, probably correctly, that unkindness is not inside that restricted range of circumstances where variations have little effect.

In warning against unkindness, Caplan stumbles into the area of parental attitudes and beliefs that help determine both parent and child behavior. Unkindness is not necessarily defined in the same way by people inside and outside Caplan’s circle. In Michael Pearl’s book, “To Train Up a Child”, Pearl and his wife propose that four-month-old infants who cry or resist a parent should be whipped with a willow switch, or, if no switch is available, a length of plumbing supply line is excellent for the purpose. The Pearls believe this is a kind thing to do, because a child who does not learn early to obey his parents will also defy God and be condemned to Hell. Naturally, if you take this view, you’ll consider it far kinder to save your child from Hell, at the expense of a few minutes of pain, rather than to let him roast for Eternity, just as some of us might think the pain of an immunization is kinder to cause than allowing the risk of death by tetanus.

Caplan’s advice to “be nice”, although undoubtedly valid, is in its vagueness a clue to the lack of real information and thought that went into this book. I can’t expect anyone to read on forever, so let me just address one example of incompleteness here: the discussion of behavior genetics. Caplan provides a handsome equation on pp. 73-74, describing the fraction of developmental variance explained by heredity, the fraction explained by shared family environment, and the fraction explained by aspects of family environment that are not shared by the family’s offspring. Very nice, but what is missing, and quite important, is the variety of interactions between heredity and environment. An example of such an interaction would be the custom in some parts of the world of giving available food to males in a family, and restricting females to whatever food is left over. This is particularly relevant to protein sources. The genetic factor, maleness or femaleness, determines how much food is consumed when resources are limited; the genetic factor thus causes an environmental effect that has a direct influence on development of both body and mind. In this type of interaction, the genetic characteristic actually evokes a response from the environment because of social attitudes.

Very well, you say, but I don’t feed my son more than I feed my daughter. What does this have to do with my parenting? What it has to do with the parenting of people who are likely to read this review is that discipline techniques, to work well, need to be matched to children’s biologically-determined temperaments, such as tendencies to explore or to withdraw from new things, or to have a generally positive or generally negative mood. (By the way, I expected Caplan’s interest in biological factors to lead him to temperament, but apparently it didn’t.) An example of research in this area is Kochanska, G., Aksan, N. & Carlson, J.J. (2005). Temperament, relationships, and young children’s receptive cooperation with their parents. Developmental Psychology, 41, 648-660.

Another important point about interactions between heredity and environment has to do with age. It’s so easy to assume that whatever is true about “nature” and “nurture” in infancy will be true in childhood, adolescence, and so on. But this is not the case, and statements about heredity and environment need to consider events across the developmental trajectory. Take for example Williams syndrome, a genetic disorder that has obvious effects but is not very debilitating. Adult Williams syndrome individuals have wonderful language abilities and are extremely sociable, although because they have little social anxiety they may make things awkward for others. So what would you think they were like as babies? Friendly, early talkers? No, they are not at all the way you would expect them to be. They have horrible colic, which lasts longer than with most babies and makes them extremely irritable. When they begin to recover from the colic, rather than laughing and smiling, they stare at people as if longing for eye contact. They don’t even begin to talk until about age two, rather than the typical 12 months or so.

These are just a few examples of the failings of Caplan’s book, which may have started as a desirable antidote to Amy Chua’s drag-them-by-the-hair philosophy, but which has ended up omitting too many details to provide useful arguments. I notice that the Wall Street Journal review admires the book for its omission of abstruse psychological concepts; that’s okay with me, but I do want to see evidence and reasoning that are relevant to claims, not just evidence that “shows the flag” but does not actually speak to the premises of the book.

[Disclosure: This review was written at the instigation of my son. Now may I play with the grandchildren?]