Fairly interesting post, especially for those who weren't aware that hallucinogens sometimes cause minor hallucinations indefinitely, but arguments like the conclusions section are why we need to, as a society, beat "the plural of anecdote is not data" into every single person's head.
Some people have a negative reaction to LSD. Some people don't. Some people have a positive reaction to LSD. Some don't. What's missing, crucially, is the only useful information- how many of each there are. There is nothing else that is relevant! If "some guy had a bad time" was a robust argument for "never ever do this thing", it would be a bad idea to do anything ever.
> Reading reddit anecdata about how LSD has changed people’s lives for the better - and the army of comments that dogpile on, agreeing - makes me upset at their naiveity and harmful promotion
It is indeed naive to think that some random guy on the internet talking about their experience is useful. Now how did we decide it was "harmful promotion" again?
The tail risk is real though. I do hope someday we’ll have accurate estimates of those risks. In the meantime it seems pretty clear to me that the expected value of these drugs is negative. I think it was Charlie Munger who said “in the best case scenario you feel good for a few minutes. In the worst case scenario it ruins your life.”
This is true for anything though. For example: I dance (professionally), surf and snowboard. All three of them have caused me injuries, though only one of them left me with something permanent (very minor, more of an annoyance thankfully) and people have gotten extremely serious injuries and even death from all of them. Should we stop doing them recreationally because of possible risks?
If the dance involved slamming your head off the pavement repeatedly until you start smelling colour then yes we would stop that form of dancing.
That's why we have things like alcohol and don't have things like heroin. It's all about risk vs reward. Heroin makes you feel much nicer than alcohol but it's almost certain to fuck your life up so we don't do it.
The head slam may be cooler than interpretive dance, but we don't do it because it gives you brain damage.
> That's why we have things like alcohol and don't have things like heroin.
No, by your logic, alcohol would have been banned long before any other mind-altering substance. A multitude of longterm studies[1] have borne out the fact that most illegal drugs are safer than alcohol on multiple metrics[2]. Contemporary drug policy is based on culture and superstition, not science and medicine.
I mean maybe? Depends on how willing you are to trade small risks for small benefit. But we have no real bounds on upsides or downsides, and no real probability of any of it. I don't know about you, but my prior in situations such as these is that if a strong effect in either direction were likely, we'd have better data.
Munger should be realistic. In the best case scenario, the drugs trigger metabolic changes that let your body deal with incipient cancer or some other significant disease.
> hallucinogens sometimes cause minor hallucinations indefinitely
The author definitely did not think those were minor side-effects.
> What's missing, crucially, is the only useful information- how many of each there are. There is nothing else that is relevant!
the evidence that a bad trip can potentially cause permanent damage is useful and relevant.
> If "some guy had a bad time" was a robust argument for "never ever do this thing", it would be a bad idea to do anything ever.
"some guy got permanently fucked up doing something" is a strong argument for me to very carefully consider the potential benefits of "doing something" and whether they are worth it; in the case of LSD, for me personally, the benefits don't outweigh the potential downsides, and anecdata is more than enough for me to make this decision.
> the evidence that a bad trip can potentially cause permanent damage is useful and relevant.
I would argue that this is not true in isolation! I guess it's not technically wrong, in that a new article is not 0 information, but without an attached probability there isn't much to learn. Frankly, some of the symptoms the author experienced are typical of HPPD, and some are not. Correlation isn't causation, and one datapoint absolutely does not show so much as correlation.
You almost certainly partake in activities with a chance to go far more wrong every day! In fact, if the mere possibility of something going wrong is enough to dissuade you, then you shouldn't have needed the article in the first place- there's always a chance things could go horribly wrong!
To be clear, I'm not saying LSD has no risks. The author got very unlucky, but not as unlucky as others who have been documented in the literature. My pitch is not even that those risks are mostly worth it- for many they probably aren't. What I'm saying is that the author's experience is every bit as useless as the people who constantly hype LSD up. The conclusion is "Fuck you, Jobs"- can you say with a straight face that the author wouldn't have concluded the same way even if no one else in the world had ever had a bad experience on LSD?
> but without an attached probability there isn't much to learn.
But there is an attached probability implicitly. I can make some common sense assumptions about
the number of people that have ever done LSD and get a lower bound on the risk (with a huge
confidence interval, but still).
Of course a chance of say 1 in ten millions is really small,
but if the perceived benefits of doing LSD are also small, then that might be enough for me to make up my mind.
> Frankly, some of the symptoms the author experienced are typical of HPPD, and some are not. Correlation isn't causation, and one datapoint absolutely does not show so much as correlation.
This is a good point, I have to decide for myself whether that person is trustworthy or not,
and whether their experience is relevant to my case.
But if I decide that I can trust their evidence, then I'm going to factor it into my decision, even
if it's only one datapoint. If in the future a controlled study on the risks of doing LSD gets published, I can reevaluate my position then.
> You almost certainly partake in activities with a chance to go far more wrong every day! In fact, if the mere possibility of something going wrong is enough to dissuade you, then you shouldn't have needed the article in the first place- there's always a chance things could go horribly wrong!
Yes, I partake in many activities that have a nonzero chance of harming me temporarily or permanently.
But with each of those activities, I've decided that the potential benefits outweigh the risks.
And mind you, I did not need to read extensive studies with attached probabilities do decide that
crossing the road or swimming in the sea is worth the risks.
This touches on another disagreement I have with your original response. We don't live our lives
constantly evaluating the risks based on peer reviewed studies with attached probabilities.
We rely on common sense and our lived experiences as well as other people's experience that we learn about.
The plural of anecdote is not data, but together with a healthy dose of common sense, it can provide
acceptable heuristics for evaluating risks and making decisions.
I think I'm making two assumptions here, and they explain the bulk of our disagreement.
First my intuition here is that in most situations, we should expect any specific example brought to our attention to be actively misleading. Out of thousands of stories about LSD, this one happens to be on the HN frontpage- should we not expect it to be unusual in some way?
Second, I think I assume that for any given activity, there exists at least one story about it that is very bad. So finding such a story shouldn't swing your judgement too much. It isn't new information!
To put it another way, we can't possibly be using the pair (average case gain, example very bad outcome) to judge if things are worth it. When we decide to cross the road, we are drawing on our extensive record of safely crossing roads, as well as our highly accurate model of the road (I would feel much less safe crossing a highway at rush hour than a suburban road at 1am) to judge the situation. If we just drew on benefit (get to the other side of the road) and potential cost (get hit by car), there would be no difference between crossing at an intersection and jaywalking. Clearly there's a probability component which is not drawn from the ether, but instead from our internal model of how the world works. I think that in your argument, this would be the "healthy dose of common sense".
In the case of drug effects, or in medical interventions overall, we don't have that benefit. Here anecdotes and common sense commonly steer us wrong- that is the undeniable conclusion of decades of disappointing research.
My specific disagreement with the way the author framed their point, which prompted my response, was how they felt that people who have had good experiences and talked about them were "naive", because the author themselves had a bad experience. I take the perspective that those people are no more wrong than the author- both are generalizing entirely from their own experiences. But your perspective would seem to argue that those people talking about how LSD is great are correct! After all, many psychonauts have tried LSD many times and come out fine, and have seen others do the same. They are reasoning from anecdotes and a healthy dose of common sense.
It may well be the case that for most peoples' risk budgets, LSD isn't worth it. The author obviously things so. But at no point do they make that argument! The sum total of the post is "Some people anecdotally say they get great benefit from LSD", "I had extremely negative effects from LSD (as do some other people)", followed by "therefore LSD is definitely not worth it for anyone". Huh? It's totally understandable why someone who went through what the author did would feel this way. It would be entirely unreasonable to expect someone bitten by a shark to couch their very real and very bad lived experience with the statistics on shark attacks. Hell, there's a legitimate point to be made about how the spiritualism and hype around psychedelics mean that positive anecdotes are treated as data while negative anecdotes are brushed under the rug. But the author, however understandably, did not make that point.
I grew up being warned that “it only takes one time” and you can fry your brain.
Then I read on HN that was BS. Flashbacks don’t happen, trips can be made safe, etc.
I’m actually glad to finally see this LSD etc promotion challenged. Seems like what I heard growing up was more true than what I’ve read on HN up until today.
But that's exactly what I'm talking about! In no sense does this article "challenge" random people on HN, because there's nothing to challenge at all. Some people did say they liked LSD, which is as true as it ever was, and other people say they don't like LSD, which is also true. Anyone going from "well I heard it's ok" => "flashbacks don't happen" was wrong before this story was ever published, because the plural of anecdote is not data.
I would encourage most people to learn to drive a motor car, and get a driver's licence. I wouldn't tell them that it's perfectly safe, because well, it isn't. It can be made _safer_ but won't be risk-free. Nevertheless, "don't try it" is not a good conclusion to draw from the existence of that risk.
Similarly, the thinking that LSD or other recreational pharmaceutical must be either entirely bad or entirely safe, the "bad outcome does not actually happen" or "it only takes one time" is binary, black or white, absolutist thinking. The truth is somewhere in the middle.
And I think that most participants know that it's an "extreme sport" a "daredevil thrill" like skydiving or a roller-coaster. A roller-coaster that never gains velocity or height would be pointless and boring.
Some people have a negative reaction to LSD. Some people don't. Some people have a positive reaction to LSD. Some don't. What's missing, crucially, is the only useful information- how many of each there are. There is nothing else that is relevant! If "some guy had a bad time" was a robust argument for "never ever do this thing", it would be a bad idea to do anything ever.
> Reading reddit anecdata about how LSD has changed people’s lives for the better - and the army of comments that dogpile on, agreeing - makes me upset at their naiveity and harmful promotion
It is indeed naive to think that some random guy on the internet talking about their experience is useful. Now how did we decide it was "harmful promotion" again?