The suicide attempt was before he turned to hard drugs, and I did explicitly say that I recognise my anecdote is an extreme case.

This. The very vast majority of cannabis consumers will never touch hard drugs, and have no desire to even consider it.

I don't know if the "gateway" concept is true or not but this fact (if true) doesn't dispute it. If you imagine hard drug use as a funnel with an optional step at the top - marijuana, with a very low conversation rate, the fact the conversation rate is low doesn't mean it's not the top of the funnel.

All of which would equally apply to alcohol or tobacco. And the data to support such a link for alcohol or tobacco is apparently just as strong, if not stronger.

But this whole line is a canard. The question is, would prohibiting any or all of those help anyone? Does prohibition help or not?

Because prohibition is the one part of the equation that a debate could actually have any affect on.

But I think it is quite clear why so much time, energy, and dubious argumentation is spent avoiding that question.

The consequences of a statement are irrelevant to the truth of that statement.

My point, as I wrote, is that it may or may not be true that marijuana is a gateway drug even if few marijuana users actually go through that gateway.

Everyone who has ever crashed in a car first entered a car.

Do you dispute that getting into a car is a gateway to car accidents?

Getting into a car is necessary for having an accident, but it is insufficient.

The "gateway drug" argument is that any amount of cannabis is sufficient for becoming a skin-popping fentanyl junkie, but it is not necessary.

> The "gateway drug" argument is that any amount of cannabis is sufficient for becoming a skin-popping fentanyl junkie

That's an extreme, and not representative of what most people mean when they say "gateway drug".

My understanding of the gateway drug argument is this: someone who starts with a seemingly-innocuous mind-altering substance is more likely to end up wanting more and going for harder stuff than someone who has never used any drugs at all. By normalizing the light drug, some percentage of the new users will end up addicted to hard drugs that wouldn't have otherwise.

I'm not familiar enough with the evidence to have an opinion on the accuracy of this argument, but it's not helpful to take the most idiotic framing of it and attack it, even if it is the framing with the greatest meme value.

The correlation is that those who use "hard" drugs have often also previously used other drugs, including alcohol and prescription medication. Flipping that around to say the using cannabis leads to injecting heroin is the classic post hoc ergo propter hoc logical fallacy.

Perhaps the correlation is a result of cannabis being illegal in your country: if you break one law (cannabis use), the barriers are down and you may as well be hanged for a sheep as a lamb and explore stronger remedies for your troubles.

So the questions that need to be investigated are these. Is there in fact any kind of correlation? If there is a correlation, is it causative? Of course, no true believer needs these questions answered, which is why the research into these things in not legal in countries where prohibition enforcement and punishment is big industry.

No I'm agreeing with it to highlight its uselessness as a thought.

Is caffeine also at the top of the funnel? How about melatonin?

The fact that most heroin users have used caffeine previously does not mean that caffeine is a gateway drug to heroin, or at all correlated with hard drug use. The same applies to cannabis.

yup. It's an interesting story, but the article is about generalizable impacts of cannabis, and this person in the story is certainly not an example of that. Granted, it was acknowledged that this was an extreme case.