Analogies are analogies, not identities. The point I was making with my analogy was that any course of action is likely to have a number of consequences. If you are trying to work out what is the best or least worst option you have to take that into account rather than pick out one and ignore the others.

If you focus exclusively on the goal of reducing the likelihood of young people getting hold of drugs of various sorts (including tobacco and alcoholic drink) from people supplying them illegally, then a policy of providing them directly could make sense. But that would involve ignoring the other consequences of doing this, including what would be interpreted as official encouragement to use the drugs supplied. (My second paragraph was tongue in cheek in this respect.)

The same kind of double effect has to be balanced in the case of schools providing birth control to "middle school kids". On the one hand it can be argued that it increases the likelihood that kids having sex will use contraceptives/prophylactics, reducing the chance of pregnancy or or STDs. On the other hand it can be argued that it might serve to increase the pressures to have sex prematurely, which in practice much of the time woudl be likely to be without making use of effective contraceptives/prophylactics.

There isn't any obvious commonsense way of knowing how the balance of advantage works out between those two consequences. That means people rely on their gut instincts and on their personal experience.

I don't find it too amazing that TRUBIT's son conmes to a conclusion on this basis which TRUBRIT finds amazing - he's in a different situation, seeing different things, and making his judgment on that basis.

And I've no idea where the truth lies, and I doubt if any of us do. Perhaps there has been research that would make the balance of advantage clearer. There clearly needs to be that kind of research as a guide to policy.