"What makes them more capricious than DUI limits for alcohol for example?"
Because if you look at the NHTSA data, it's very clear that the risk for someone driving at .08 is about 5x the baseline risk. If there was similar research showing that driving at 5ng/ml of marijuana had the same 5x increased risk then it would make a lot of sense. But that is not in fact the case. What they are essentially saying is that it's ok to drink and drive until your risk of killing someone is 5x increased, but it's illegal to smoke and drive even if your risk factor is only 1.1x or whatever.
Comments
"What makes them more capricious than DUI limits for alcohol for example?"
Because if you look at the NHTSA data, it's very clear that the risk for someone driving at .08 is about 5x the baseline risk. If there was similar research showing that driving at 5ng/ml of marijuana had the same 5x increased risk then it would make a lot of sense. But that is not in fact the case. What they are essentially saying is that it's ok to drink and drive until your risk of killing someone is 5x increased, but it's illegal to smoke and drive even if your risk factor is only 1.1x or whatever.
http://www.nhtsa.gov/people/injury/research/AlcoholHighway/2... (See figure 2-6)