I’ve been reading the blog Free Range Kids by Lenore Skenazy for a while, and her book by the same name has been on my “to be read” list for a while. The point of the blog (and I assume the same is true of the book):
At Free Range, we believe in safe kids. We believe in helmets, car seats and safety belts. We do NOT believe that every time school age children go outside, they need a security detail. Most of us grew up Free Range and lived to tell the tale. Our kids deserve no less. This site [is] dedicated to sane parenting.
I agree with Skenazy’s premises, and usually agree with her conclusions. But I also quibble with what seems to me to be an important rhetorical flaw I’ve been seeing in some of the illustrations in her blog posts. Those illustrations, at times, contain no aggregate risk and utility analysis.
I’ll explain what this means, but first here’s why I think this matters: when you’re having an argument, you’re probably not winning when you have to say “wait, that wasn’t a good example; how about this”. And I think that in her blogging, in addition to plugging her book, Skenazy genuinely hopes to give some ammunition to her readers who run up against people saying nutty things about parenting. But considering only part of the aggregate-risk-utility test leads to bad examples (where I suspect there are really good examples to be given).
Here’s what I mean by “aggregate-risk-utility test”. Suppose you’re considering whether to try to safeguard your kids against some kind of harm. The utility to you of not taking that precaution includes the money you’ll save, the other cool things you can do with the time you would have spent on the precaution and the intangible benefit of not training your kid and yourself to be scared all the time. That utility has to be weighed against the probability that skipping the precaution will result in harm befalling your children multiplied by the magnitude of the harm. This calculation has a storied past in philosophy* and law,** and it’s something most people understand without being told.
Today, Skenazy, in deflating the idea that you should snap a picture of your kids each time you leave the house with them, says:
Not that we shouldn’t have an up-to-date photo of our kids. That does make sense. But to make this a part of one’s DAILY routine, like flossing, is to assume that kidnapping is as likely as tooth decay.
As usual, I agree with Skenazy’s point—the idea of taking a photo every day so that you’ll have the very latest and greatest to give to the evening news folks is pretty silly. But there are three levers in the aggregate-risk-utility inequality, and by only moving one of them (the likelihood of harm), Skenazy leaves an important question unanswered—what if the harm of child abduction is so much greater than the harm done by tooth decay that the utility cost of snapping the picture appears to be lower than that of flossing? I bet most parents have a hard time considering a question like that without imagining the worst. And when you imagine the worst, the Free Range quote above ends up reading a bit like an argument for taking that daily picture. (“You floss your teeth every day just to avoid a little tooth decay; is it really so silly to snap a quick pic to increase the likelihood that your abducted child will be returned to you?”)
I know I’m not saying anything Skenazy doesn’t know. The premises of much of her work seem to be that (a) we seem to be having a hard time assessing the likelihood that we can save our children from harm and (b) we live in a culture (CSI, the evening news) that fetishizes worst-case outcomes, skewing our assessment of what kind of harms our kids might come to, all pointing to the conclusion that we are making some bad decisions about the utility costs of our safeguarding measures. And my hunch is that the phenomenon I’m complaining about here isn’t a feature of her book, but is rather a by-product of the inevitably rushed pace of a busy person’s blogging (sort of like Matt Yglesias’s spelling).
P.S.: This blog post began as a response to a facebook status comment. After I posted a variation on this quote:
If, for some strange reason, you actually WANTED your child to be abducted by a stranger, do you know how long you would have to keep her outside, unattended, for this to be statistically likely to happen? Guess. Now guess again. Oh, forget it. The answer is 750,000 years.
a friend responded: “I think the lady defeats her arguments because she is so dismissive of things that can be common-sensical. I agree that kids need to be more independent (by far) but she seems to scoff too much to be taken seriously.” I’m not sure about the second complaint (as my grandfather used to say, scoffing doesn’t make you right or wrong—it just makes you a scoffer). But I suspected that the first point (the one about self-defeating) expressed a recognition similar to my own that I’ve shared here.
And as for common sense, if Skenazy is even half right, common sense needs to be seriously re-thunk.
* Blaise Pascal famously conjectured that believing in god costs you very little utility, while the magnitude of the risk you run by staying home on Sunday is Damnation for All Eternity, which is significant even when multiplied by the tiny risk that there is, in fact, a god who will thus punish you.
** Judge Learned Hand said that the law would only find you negligent if the probability of an injury being caused by your actions multiplied by the magnitude of that injury exceeds the utility of your actions.