If it is, then a lot of us are being fairly derelict as parents. Edwards is insistent that she is not acting out of emotion, but simply "straightforward risk analysis: a cold, hard, unemotional reading of the statistical data". So clearly the rest of us have been remiss – we've apparently failed to make that risk analysis, failed to read the data properly, and we're putting our children in danger by doing so. If I accept the article's reasoning, it would be positively reckless of me not to implement a similar policy immediately. But I don't. As I'm a man, I guess that's not surprising, but hopefully my refusal to follow Edwards' example isn't based entirely on the prejudice of my gender. The first thing to note is that Edwards, despite her "cold, hard" approach to statistical analysis, has not, in fact, restricted access to her daughters for all men. Her husband, it seems, is still allowed to look after the kids unsupervised. But why is that? As Edwards notes, most children are abused by someone they know – they're more at risk from family members than strangers. Abusive fathers are a well-known phenomenon. And her rule is specifically designed to remove the necessity of trying to judge men's characters before deciding to allow them to be alone with the children: she doesn't want to make a "moral assessment" of individual men to determine their likelihood of being an abuser. Why would her own husband be exempt? The data she's working from indicates that he is as much a threat as any other man. It seems the ban isn't as "unemotional" as Edwards claims: clearly she does allow her feelings for at least one man affect the implementation. Furthermore, does Edwards apply strict statistical analysis to all potential threats? If statistics showed a greater prevalence of crime among members of a particular race, would she factor that into her risk assessments? If shown that violence is more common within a certain socio-economic class, would she make sure her children avoided members of that class?

But, putting these musings aside, let's look a little closer at the basis for this specific policy. It is a rule designed to protect children, which is obviously a noble goal. But the risk analysis takes into account only the protection of children: Edwards has started from the premise that it is a parent's job is to protect their kids from predators, and leapt from there to the belief that this is a parents-only job. It's actually pretty easy to protect children from abuse: all you have to do is keep them locked up without contact with other human beings until they turn 18. Nothing could be simpler. We don't do this because we recognise that, as important as keeping our children safe is, safety isn't a child's only requirement. Children need to grow, to experience life, to learn how to be a good, capable, confident adult. The guidance and nurturing of our children is our responsibility, every bit as much as their protection. Edwards obviously doesn't keep her daughters locked up, because she knows she knows there has to be a balance struck between safety and the experience and enjoyment of life. I believe she has struck that balance poorly. I believe that in looking at the statistics and reasoning that the best response is to shield her children from all men (almost) without exception, she goes too far in her completely understandable desire to protect them. The things we teach our kids will stay with them their whole lives. We do them a disservice if we teach them to live in fear.