Jon Caulkins has a first-rate essay in National Affairs on what the numbers actually say about the dangerousness of cannabis.
The real trouble is not that marijuana is more or less dangerous than alcohol; the problem is that they are altogether different, and the comparison is simply unhelpful in informing the debate over marijuana policy. The country is not considering whether to switch the legal statuses of alcohol and marijuana. Unfortunately, our society does not get to choose either to have alcohol’s dangers or to have marijuana’s dangers. Rather, it gets to have alcohol’s dangers — modulated perhaps by higher or lower drinking ages and higher or lower taxes — and also marijuana’s dangers — modulated by how legalization or prohibition affect prices, product variety, marketing, and usage.
Instead of comparing the harms of marijuana when it is prohibited to the harms of alcohol when it is legal, an intellectually honest marijuana-policy analysis ought to compare all possible harms under marijuana prohibition to all possible harms under legalization. And that analysis ought also to worry about indirect effects on abuse of other illicit drugs, alcohol, and tobacco — outcomes for which the current evidence is scant, divided, and discouraging, respectively.
Destroying peoples lives over cannabis is just plain wrong. I am 59 and have been using cannabis for 42 years. I’m a husband, father, musician, electronics engineer, community volunteer and successful business owner. I vaporize cannabis (no smoke, no smell, no problems). I am in excellent health and I run 25-30 miles per week and finished a Marathon last May in 3hrs, 54mins. I am just the tip of a very large iceberg! Critical mass has been reached! How long do we have to suffer the lies and ridiculous nonsense from a government whose own studies have shown cannabis nearly harmless by comparison to prescription drugs, tobacco and booze??
Legalize, regulate and TAX!
I think that second quoted paragraph captures what's wrong with the debate, but with the data problem foremost.
In the absence of good data, we really don't want to look at "all possible harms" — people have lousy intuitions about what's possible, and the formulation "all possible" is an invitation to the kind of "1% Doctrine" thinking that has worked out so disastrously in so many areas of discourse.
We absolutely need to look at what's probable rather than what's "possible", but we don't have the information needed to do that. And the only way we're going to get some of that information is by risking major mistakes.