Glenn Reynolds makes an articulate point about paternalism.

Another — and it’s a lesson that policy wonks seem slow to learn — is that people other than policy wonks are capable of learning, and of changing their behavior on their own. Given the chance, and the information, they observe things that work and things that don’t, and adjust their lives in ways that seem most likely to get them what they want, if they are allowed to do so. This means that projections of catastrophic future ills are usually wrong, as people learn from experience.

I think that perhaps the most important issue to which this might be applied today is that of obesity.

When I was in St. Louis helping take care of my Dad after a fall a couple of weeks ago, I took a walk in Forest Park. I saw a woman doing strenuous sit-up exercises and I remarked that they looked difficult. She was just finishing, got up, and joined me for walking. She was eager to talk about her new commitment to fitness.

She described her exercise regimen, and I asked her about diet. She said that she now eats mostly steamed fish and fruits and vegetables, which sounds to me like a Dean Ornish approach rather than one of the more dubious fad diets.

She said that she had lost 12 pounds in three weeks, and she still had a long way to go, but her goal was to lose weight slowly and keep it off, not lose it quickly and gain it back. Again, this is consistent with what I have heard is the best medical advice.

What struck me most of all was that this well-informed approach for fighting obesity was being undertaken by a middle-aged African-American woman, unemployed (she was going to start a job as a medical courier the following week), married to a janitor. It seems to me that if someone of that demographic knows how to fight obesity on her own, then maybe going after Big Food with lawsuits is pretty misguided.

For Discussion. Hal Varian once sent me a note about one of my essays in which he said that rather than a distinction between “haves” and “have-nots,” the real digital divide is between “information wants” and information “want-nots.” Is there a paternalistic solution for this?