In general it is a bad idea to give space to bad ideas.

But some bad ideas receive so much attention that a response is necessary. Ann Crittenden's new book is in this category.

The problem begins with the TITLE: The Price of Motherhood: Why the Most Important Job in the World is Still the Least Valued (Metropolitan Books, 2001). Strictly speaking, motherhood is not a job at all. It is a vocation. The family is not just a special case of something else, like a job or a contractual obligation. The all-important human endeavor of raising children, creating relationships and building up families needs to be described and defended on its own terms. Describing motherhood in the language of commerce and employment contributes to the commercialization of society in general and the family in particular.

So, too, does Crittenden's proposal to pay mothers to take care of their children. This proposal has generated powerful, visceral and mixed reactions. It is not hard to see why. Motherhood is valuable, and we find it appealing to reward motherhood. But at the same time, we instinctively recoil from the idea of paying mothers. People sometimes have trouble articulating the underlying problem, but I think this is a key issue: “Mother” describes a relationship, not a job. We intuitively know that you cannot pay someone to be in a relationship, and still have the relationship be authentically personal. Nor can you pay someone to do the work of being in a relationship for you.

Building up relationships is at the heart of the maternal vocation. Paying people to build relationships can seriously change the character of those relationships. The market system of exchange works well among strangers with whom you have no personal relationship. Indeed, the market is a wonderful institution precisely because it induces people to cooperate who might hate one another under other circumstances. But the family is not an impersonal network of strangers. It doesn't make sense to replace the most personal institution in the world, the family, with an institution designed for arm's-length transactions among strangers. Of all the relationships in our lives, surely, we want our mothers to take care of us because they love us and not because they are paid to.

Furthermore, this proposal to pay mothers suffers from the same defect as the old-fashioned, discredited welfare system. Because the payment is made directly to mothers, fathers and mothers have less incentive to get married and stay married. Crittenden herself acknowledges as much when she emphasizes the benefits of mothers being more independent of fathers. She quips at one point that many women are “just a husband away from poverty.” She even lets it slip out in her glowing discussions of Scandanavian family policies that almost half of all children in Sweden and Denmark are born to single mothers, in contrast to a third of all births in the U.S.

It is not accurate to describe mothers as ‘dependents.’

We know, from mountains of social-science research, that children do better in two-parent, married-couple households than in any other type of household. Children of intact marriages do better on virtually every measure than those of divorced parents, stepparents, never-married parents or cohabiting parents—health, schooling, emotional stability, propensity to remain drug-free and, ultimately, the probability of getting and staying married themselves. Surely the state should not pursue policies that decrease the probability of mothers and fathers staying married to each other.

The claim that mothers can be made independent by these payments is an illusion. Raising children is too much for one person to do alone. Someone must take care of the baby, and someone must take care of whomever is taking care of the baby. Crittenden's suggestion that the government pay mothers simply masks this underlying dependence. Mothers would be dependent upon the government, and the vagaries of politics, instead of upon their husbands, just as today's single mothers depend on the competence of their child care providers, and the good graces of their employers.

It is not accurate to describe mothers as “dependents.” Mothers must be interdependent with someone in order to bring up their children. The real questions are: Upon whom shall we depend? And with whom shall we interact? The modern world increasingly answers that women shall depend upon impersonal institutions rather than upon individual spouses and fathers. Mothers shall interact with no one in particular.

Finally, the primary emotions that come through this book are fear and anger—fear of spousal abandonment and anger at gender-based unfairness. These are simple, gut-level emotions that are easy to stimulate. People are tired of going through life feeling cheated or frightened. Organizing one's life around these emotions can never make a person happy. I believe this is the core reason why old-fashioned feminism no longer appeals to young women.

Love is the alternative to fear and anger. Using love as an organizing principle for one's life complicates things quite a bit. We can never be completely sure that we're doing it right. Our emotions are less reliable guides for action. We sometimes have to admit to our family members that we were wrong. But this approach at least has the potential to make a person happy.

Motherhood describes a relationship, not a job title. The primary role of the family is not to transfer resources from big people to little people, or from male people to female people. The primary function of the family is to build up relationships. The proper posture for family members to adopt toward one another is the posture of love. Public policies built around fear and anger can only inhibit the development of that posture. Only love can make people happy.

Jennifer Roback Morse is a research fellow at the Hoover Institution and the author of Love & Economics: Why the Laissez-Faire Family Doesn't Work.