> We very greatly vary in our regulation of those substances. And of everything, for that matter. Alcohol's sold in corner stores, Percocet is only (legally) available through a pharmacy with a prescription. Owning a concealed firearm requires a permit. If we're going to legalize 'harder' drugs, where should they fall on that spectrum? Should we allow people to buy heroin at the corner store? Maybe! Should advertisers be allowed to buy huge billboards advertising their brand of crack? Possibly. We should evaluate what's going to provide the most overall utility for us as a society. If the answer is to legalize cocaine but disallow advertising of it, then that's what we should do.
Three points on this:
1. When we talk about making drugs legal, we mean for recreational use. It's already the case that all drugs with a "demonstrated medical use" (let's ignore for a moment how that gets decided) are legal to use for medical purposes, and additionally, most of the laws have as a affirmative defense that your actions were medically necessary in some way.
2. The two legalized recreational drugs that we have (alcohol and tobacco) are sold in a variety of ways by jurisdiction, and our debate about removing the "only for medicine" requirement doesn't require that we resolve the debate about how we're going to realize that change in a uniform way. It's certainly not the most important question for if we legalize wider recreational drug use or not.
3. "Maximum utility for society" is a notoriously hard way to make a decision, because we don't necessarily agree on what the metric is. Taking my hypothetical from before again - is it of maximum utility to prohibit gay marriage because it encourages straight couples to have more kids? Well, that's incredibly hard to assess, even once you pin down several of the subjective aspects. Instead, courts have been deciding the merits based on the impact this has to freedom and how consistently that reasoning is applied across similar cases. I greatly approve of this method of decision making. It's generally better to start from first principles of the things an individual should and shouldn't be allowed to do on a whim, by getting permission from his peers, only with permission from a specific person the action is happening to, or never. It's similarly reasonable to look at how people raising objections apply those same objections to similar cases - if they're consistent, then perhaps they have a real objection; if they're not, they need to explain this either by differences between the two cases or concede that's not their true objection.