1As with many life events, divorce is a subject for discussion and debate involving a broad range of disciplines in the humanities and social sciences. Economics is not generally at the forefront of these disciplines; it is lawyers and psychologists who are first in line to accompany conjugal trajectories. Demographers are also widely called upon to examine changes and variations in the phenomenon, providing the necessary basis for formulating possible interpretations of the reasons for these changes, which occur in time, space and, naturally, the social space. Sociologists and economists are part of this second circle of experts interested in the causes and, above all, the consequences of divorce.

2The article by Cécile Bourreau-Dubois and Myriam Doriat-Duban is unquestionably a useful contribution in this respect. It provides a basis for an actuarial analysis aimed at judging the nature of this “risk” and envisaging the grounds and methods for covering it. The article proposes a model that could probably be converted into a formula at some later stage.

3Considering divorce as a “risk” is not new. For our part, we have proposed the notion of “solitude risk” (Martin, 1993). Depending on the angle chosen, divorce can be seen as a risk in the insurable sense of the term, as is the case here, or in the sense of a threat. The latter raises the question of the social construction approach to risks, including controversies about the meaning to be attributed to covariations, or disagreements over the attendant independent and dependent variables.

4One of the most significant examples of this battle concerns the effects of divorce on children. A considerable share of the scientific literature in the social sciences, particularly in the United States, has focused on this question. However, an attentive review leads to the identification not just of major disagreements on the idea that divorce is an event whose effects work mechanically to the detriment of children’s well-being, but also of disagreements over the systems used to test such a causality. For example, many authors have questioned approaches that tended to merely compare (even by matching them) children raised and socialized in households of married couples with children whose parents have divorced. The independent variable, likely to be severely neglected, concerns conflict. Without going further into this example, we can clearly understand the nature of the discussion and the importance of diversifying the situations and cases to be compared, such as those of children socialized in couples who have remained married despite permanent and violent conflict. Consequently, we have proposed the idea of assessing the “risks of non-divorce” for children in situations of conflict (Martin, 2007).

5In short, while the economic dimensions of divorce can potentially be informed in terms of its direct and indirect costs, it is more difficult to address the medium- and long-term consequences or the effects on children. The risk, when keeping to an “all other things being equal” standpoint, as is the case with part of the North American literature, is to generalize the “depreciation of the human capital of future citizens” caused by divorce to a large population of individuals, children, teenagers and young adults, who may nonetheless be capable of converting the ordeal of divorce into a strength for their future existence and human capital. The question to be asked, then, is how thinking of the kind proposed here could integrate in its bundle of variables this phenomenon of resilience or of conversion of the ordeal into an alternative strategy to avoid the silent mechanics of what are considered as normal or successful trajectories, at a given time and a given level in the social space. Another example illustrates the complexity of the variables in question: the fact that young people having experienced their parents’ process of divorce and lived with their mother as a lone parent leave education earlier on average than those socialized in a stable couple is often seen as a possible sign of “academic failure” or as a penalty linked to divorce. But that would be to neglect the fact that this effect may also correspond to a strategy, desire or need: that of becoming independent at a younger age and entering the job market sooner, even if this means shortening an educational trajectory that in other circumstances could have been longer.

6A number of other points could be discussed here, including the way the authors consider divorce as a particular risk because it results from the choice of at least one of the protagonists. Viewing divorce in terms of choice is once again highly questionable, and it naturally echoes the idea that divorce is caused in many cases by a lack of investment, effort and cooperation on the part of the protagonists. People often say that divorce numbers have increased because adults “throw in the towel” at the first sign of a problem. From that perspective, we could even ask ourselves whether we need to think about socially protecting individuals from these consequences, other than asking them to take out an insurance policy, as they would for driving a car. Thinking in terms of choices here is tantamount to considering as negligible all the factors that contribute to divorce, far beyond the question of choice. These include the links (co-occurrences) between unemployment and divorce, between loss of social status and divorce, or between everyday living conditions, the gender division of roles or achieving a balance between family life, working life and divorce.

7In other words, while it is very useful, and even necessary, to discuss the conditions in which divorce could be considered as a social risk, it is clear that we still have a long way to go if we are to avoid creating measures that are incapable of compensating for the inequalities generated by these ordinary accidents of contemporary life trajectories. Among the points to be pursued in this discussion, let us conclude by mentioning the indicators of well-being (or ill-being), which, looking beyond purely economic issues, could be taken into account in the design of measures to accompany divorce.