It is ok and not at all inappropriate to seek help here. I myself suffer from depression, so I do know the kind of pain you`re in- right now, as I write this post, I nearly weep tears of sorrow for the fact that you hurt so much right now, and my heart flies out to you.
Please do not kill yourself!
Although it may seem very far away right now, the pain can go away, and you can be happy again.
As for an immediate plan of action I would recommend the following things:
first, numb the pain so you do not get worse. Thats not a longterm-solution, its just until this cry of help does help you. What worked for me was watching old TV-Series and Starcraft-2 casts.
second: I certainly do not know your reasons why you do not wish to seek seek professional assistance, but I do understand this reluctance of yours. I myself needed over a year to make the call; a time which I now wish to be much shorter. So i ask you to reconsider: professional help is the fastest form of sustainable relief you can get and is probably the best method to minimize the total amount of suffering. For this, I would find it fantastical if you would not only talk about the reasons why you hurt, but also about the reasons why you do not wish to seek this form of help.
third: Should these feelings suddenly intensify and you really really need someone to talk to at, say, 02 00 am, it is helpful to have saved the number of a suicide hotline into your phone (or some other immediate form of communication, where you do not have to wait for long, thoughtful answers. Set up a IRC-chat.)
It is ok and not at all inappropriate to seek help here.
I think it is. We are not suicide experts, we know nothing about how to help, and our attempts to help can easily backfire. And this is assuming that we are not being trolled, attacked, or drama queened (anyone can say that they have contributions under another account). Let us recall that pdf23ds asked for LW help with his sleep problems which did nothing, and LWers attempting to follow up on his posted suicide note apparently did not solve the problem.
The big sites like Wikipedia generally have a policy of:
not encouraging amateurs to engage in therapy-over-the-internet and other-optimization
having the admins escalate to looking up IP addresses and contacting locals
providing the concerned user with a suicide hotline number
I don’t see any reason that this should not be implemented on LW as well.
OTOH, suicidal people often have very few options, so removing one is a big cost. Lots of reports suggests that suicide hotlines aren’t actually much better than random like-minded people, though it’s possible that callers underestimate how helpful they were. And contacting locals is among the most likely strategies to backfire.
I wasn’t suggesting banning users contemplating suicide. (Do any sites do that? Is it a good idea?)
Lots of reports suggests that suicide hotlines aren’t actually much better than random like-minded people, though it’s possible that callers underestimate how helpful they were. And contacting locals is among the most likely strategies to backfire.
I want research, too, but in lieu of that, I second “lots of reports”. Also, consider this reality:
Imagine you tell someone that you’re thinking about killing yourself. Say this person decided to call 911, and you end up admitted into the mental ward of a hospital.
Unless you have very good insurance (or something has changed), they’re likely to let you out in a week. I’ve heard of this happening to countless people. It is simply too expensive to keep them in there very long.
So, a week later, you’ve:
Missed a week worth of pay (from work).
Racked up tens of thousands of dollars in hospital bills (most likely your insurance won’t pay for it all.)
Missed a week of work. Some workplaces will find a reason to fire you after that. (Not all managers care that this is illegal.)
You had to live in a place full of crazy people every day for a week.
You had to sleep in a place full of crazy people, perhaps sharing a room with one, every night, for a week.
You’ve been treated like a crazy person for a week. (Barely allowed any possessions, unable to go anywhere, living apart from friends and family, having your life scheduled by the staff, etc.)
Your new prescription may not even kick in for several weeks!
Is your situation and mood:
A. Better
B. Worse
The last suicidal person I called 911 for was not provided any services, but was charged hundreds of dollars. Apparently, they stopped by, asked him a pile of questions, he managed to avoid being taken to the mental ward, and then they charged him out the ears for the visit.
He was quite upset with me. Quite. Moral of that story: Don’t IM about suicidal thoughts while the internet connection is flaking out and then say something like “I can’t go on” and then also fail to answer the now very concerned friend’s phone call right after.
It’s enough to make me think that someone should do a study to determine whether fatalities might increase after a visit to a mental hospital.
And as for the suicide hotlines—a lot of the operators are volunteers. I wouldn’t be surprised if the number of operators who are volunteers is like 90%+. I’m sure you have to be pretty special to be a suicide hotline volunteer (nerves of steel + screening) but I doubt that most of the operators are psychologists. This could probably be verified if you were willing to look up the largest suicide hotlines and see how they source their operators. For now, a quick example of operator sourcing:
Crisis Clinic Volunteering Page:
“Crisis Line phone workers come from all walks of life and educational backgrounds.” (All educational backgrounds means lots of people who are not licensed psychologists).
It is interesting that the suicide rate for people who have just left a mental ward is over twice that of those who have just started anti-depressants, but:
A. Those who end up in the mental ward are more likely to be in that group because they actually attempted to kill themselves, or because they have a plan, whereas the anti-depressant patients and other groups of people might be experiencing suicidal ideation without an attempt or plan, or simply just depression without suicidal ideation.
B. People who plan or attempt to kill themselves are probably a lot more likely to attempt to kill themselves in the future than people who have never attempted and may not even be thinking about it.
In order to know whether admitting a suicidal person into a mental ward does more harm or more good, we’d need a study where they compare outpatient suicide rates with suicide rates in a control group. A good control group would need to consist of equally suicidal people who were not admitted into a mental ward. In order to ensure they were comparable, they’d need to track things like whether a suicide attempt was made, whether the person was on anti-depressants, etc. Also, if nobody had hospitalized the hypothetical control group, there’d be a reason for that. This reason would have to be the sort of thing that wouldn’t impact the suicide rate itself. For instance, if those people are living in a place with no legal way to hospitalize a person against their will, it might produce a great control group. On the other hand, if the control group mostly consisted of people who were not hospitalized because they are alone and have no family members or friends who care enough to call 911, this would not be a great control group.
Now I wonder if there are any places where you cannot legally be admitted to a mental ward against your will… or whether there might have been a point in the past when there was a place like this.
It’s hard to imagine that there would ever have been a place like this, but you never know.
You’re quite right that it’d be very hard to demonstrate a causal effect (and without having read the study itself—don’t have access -, I suspect the researchers didn’t even want to try).
Actually, I have no idea how that could be done in practice. For voluntary hospitalization, it would be helpful if one couldn’t be hospitalized against one’s will, but I’m not aware of a time and place where that would be the case, and don’t expect there to be any. So one can study that only outside the realm of “hospitalization-worthy” suicidality, by using patients who have been offered hospitalization, but declined it, as a control. My quick search turned up no indication of even that having been done.
And for forced hospitalization, it seems sort of impossible in principle to find a control group...
I also just found this, again I don’t have access, and unfortunately it doesn’t even have an abstract. Might be relevant, though, judging by the title.
Check for that first link on Pubmed. I was able to access the article that way.
I’m not finding this new article anywhere. The text in a Google book search where the study had been mentioned suggests that it’s probably not the type of study I’m looking for though.
I doubt this type of study exists. About the best we could do is to compare the suicide rate in an area with no legal method for hospitalizing suicidal people against their will (if that exists and the country is developed enough to keep such statistics reasonably well) with the suicide rate of a comparable area.
I looked at the results section of that abstract, but there’s an obvious reason why their correlation of an increase in suicides after hospitalization may not have Epiphany’s causal explanation.
I didn’t mean banning, just people thinking “They won’t help me, no point posting” or “Holy mackerel they’re going to call 911 on me and get me locked up” or “I would ask LW for help, but I can’t get over fretting about the reaction”.
A couple mental illness support groups I know of ban talking about planned suicides (and self-injury), mostly to avoid triggering potentially suicidal (resp. self-injurious) people. Since they point to emergency resources a lot and support non-acutely suicidal people, it doesn’t seem to be very bad, but I can’t observe the consequences.
Are these claims research-based?
The first is anecdotal. I’ve heard a lot of negative reports and very few positive ones, but there may be selection bias. Maybe hotlines are for people who are desperate to talk to anyone, so people who’d talk to me don’t benefit nearly as much.
The second… is probably wrong, come to think of it. A lot of people shriek “Never spill the beans about someone being suicidal without their consent, it could get them into arbitrary amounts of trouble, and an even suckier life won’t make them less suicidal”, but even more people shriek “Call 911 NAO” and the main evidence that the former know what they’re talking about is that they have much more personal experience.
While I agree in general, I do think such guidelines and agreements are best discussed in a quiet, separate thread, with explicit agreement of the concerned Mods, and not in response to an “emergency” call for help. In this instance, I am very glad that Slade did manage to call for help this way.
These are good points, Gwern, but the psychology industry and suicide hotlines fail a lot more often than you’d think (support for this point is included). If Slade has not tried them, they may still be useful for Slade. It is also possible that Slade has tried a few different psychologists, a few different prescriptions, and a few different hotlines without relief. Here are a few problems that one may run into when attempting to utilize the standard forms of help:
Note to Slade: You may not want to read this. This is for people who don’t have a clue why anybody would be asking here. It consists of a list of problems. It’s useful for the purpose it’s serving, but not particularly uplifting for a person in your situation. I’m writing you a separate response.
Psychologists failing to practice science. I’m sure you, Gwern, are aware how common this is but I’ll include a link for others.
Prescription drugs to treat moods not working. (See “A link for others”.)
Prescription drugs to treat moods causing intolerable side effects. (The FDA doesn’t approve drugs based on them being safe. It approves them based on whether the problem caused by the drugs is better or worse than the original affliction. I will use tardive dyskinesia, a neurological disorder resulting in involuntary body movements like repeatedly sticking one’s tongue out, as my example. Anti-psychotics can cause it.)
If Slade is gifted (not unlikely according to these numbers and Slade may have a 50% chance of not knowing (“about half of our country’s (America) gifted students are never identified”)) then they may find themselves being misdianosed with disorders they don’t have (happens to 25% of the gifted population (Citation: Misdiagnosis and Dual Diagnosis of Gifted Children and Adults) let alone the portion of the gifted population which actually shows up in a psychologist’s office).
If Slade is gifted, they might find that getting help from a psychologist who does not specialize in gifted clients is useless, frustrating or harmful. This can feel like taking a space ship to a car mechanic. Gifted people may need psychologists who have experience with gifted clients.
Suicide hotlines are non-profit organizations. As such they can end up understaffed. The following scenario does happen: Can you imagine how it would feel to call a suicide hotline and be told “Hold on a minute, I’m getting another call.” What if it kept happening over and over again throughout the call? If you look past the obvious annoyance that this is likely cause during your important conversation about this life or death matter, you’ll also see that you may begin to wonder how many other people are killing themselves while you receive the help. Total survivor’s guilt.
It is quite possible that Slade has attempted to get help using the stereotypical options and has run into one or all of these issues.
I don’t blame Slade for asking here. In the event that Slade’s main problem is that the psychologists don’t seem to get it—that Slade is gifted and they’re targeting their help to non-gifted clients, or that Slade is receiving unscientific advice from psychologists—one of the places where Slade would have a decent probability of finding these things out is by asking here.
If Slade has been failed by the ordinary methods, would you agree that the situation is important enough to “throw spaghetti at the wall” by posting on the internet asking for help, and that LessWrong would have a significantly better chance of producing a useful response than most websites?
Personally, I think the best approach when dealing with a problem this big is to talk to as many different people as possible about it—including both professionals and laymen that are either knowledgeable about the subject or generally smart. Nobody is infallible, and the more information you have, the better.
We are not suicide experts, we know nothing about how to help, and our attempts to help can easily backfire.
This sounds to me like taking the precautionary principle too far. Shouldn’t we do an expected value calculation? I imagine the folks at Wikipedia had a few incidents of people being helped with their depression and killing themselves subsequently (possibly as a result of the help they received), then Wikipedia was like “What should we do? How about a don’t-talk-with-suicidal-people policy?” without looking at the other, much more obscure cases where someone asked for help and did not subsequently commit suicide (possibly as a result of the help they received).
I seem to have read that talking to a random smart, empathetic person is just as good as talking to a pro counselor, BTW.
I imagine the folks at Wikipedia had a few incidents of people being helped with their depression and killing themselves subsequently (possibly as a result of the help they received), then Wikipedia was like “What should we do? How about a don’t-talk-with-suicidal-people policy?” without looking at the other, much more obscure cases where someone asked for help and did not subsequently commit suicide (possibly as a result of the help they received).
I don’t know. Wikipedia spends a lot less time on socializing then LW does.
No one can stop them even if it’s a terrible idea, they have nothing to lose, and nevertheless, they list their list of suicide hotlines prominently in the sidebar.
having the admins escalate to looking up IP addresses and contacting locals
How much can you narrow down my identity from a comment, I really have no idea? That’d probably only cause trouble for me, by the way.
trolled, attacked, or drama queened (anyone can say that they have contributions under another account)
There’s really no reason someone couldn’t do those things even if they did have contributions under another account. And I’m curious, what do you mean by “attack” other than trolling?
LWers attempting to follow up on his posted suicide note apparently did not solve the problem.
There is a difference between a suicide note by someone determined to kill himself and a plea for advice from someone who doesn’t particularly want to die but might kill himself all the same. Your point about attempts to help plausibly backfiring still stands, though.
How much can you narrow down my identity from a comment, I really have no idea? That’d probably only cause trouble for me, by the way.
I don’t know; it depends on the details. I know from my time as an admin on Wikipedia that we were often able to contact local authorities or family, often using IP addresses. (I imagine suicide is quite troubling too.)
There’s really no reason someone couldn’t do those things even if they did have contributions under another account.
It reduces the odds.
And I’m curious, what do you mean by “attack” other than trolling?
I think that if you can’t think of ways in which suicides, real or faked, could be used to attack a group, then it’s probably better if I don’t list every idea that comes to mind.
There is a difference between a suicide note by someone determined to kill himself and a plea for advice from someone who doesn’t particularly want to die but might kill himself all the same.
I am not an expert on suicide (which is one reason I think a LW suicide policy is a really good idea), but I understand many or most suicide attempts fail so I’m not sure there is a difference or if the difference is important.
It is ok and not at all inappropriate to seek help here. I myself suffer from depression, so I do know the kind of pain you`re in- right now, as I write this post, I nearly weep tears of sorrow for the fact that you hurt so much right now, and my heart flies out to you.
Please do not kill yourself!
Although it may seem very far away right now, the pain can go away, and you can be happy again. As for an immediate plan of action I would recommend the following things:
first, numb the pain so you do not get worse. Thats not a longterm-solution, its just until this cry of help does help you. What worked for me was watching old TV-Series and Starcraft-2 casts.
second: I certainly do not know your reasons why you do not wish to seek seek professional assistance, but I do understand this reluctance of yours. I myself needed over a year to make the call; a time which I now wish to be much shorter. So i ask you to reconsider: professional help is the fastest form of sustainable relief you can get and is probably the best method to minimize the total amount of suffering. For this, I would find it fantastical if you would not only talk about the reasons why you hurt, but also about the reasons why you do not wish to seek this form of help.
third: Should these feelings suddenly intensify and you really really need someone to talk to at, say, 02 00 am, it is helpful to have saved the number of a suicide hotline into your phone (or some other immediate form of communication, where you do not have to wait for long, thoughtful answers. Set up a IRC-chat.)
I think it is. We are not suicide experts, we know nothing about how to help, and our attempts to help can easily backfire. And this is assuming that we are not being trolled, attacked, or drama queened (anyone can say that they have contributions under another account). Let us recall that pdf23ds asked for LW help with his sleep problems which did nothing, and LWers attempting to follow up on his posted suicide note apparently did not solve the problem.
The big sites like Wikipedia generally have a policy of:
not encouraging amateurs to engage in therapy-over-the-internet and other-optimization
having the admins escalate to looking up IP addresses and contacting locals
providing the concerned user with a suicide hotline number
I don’t see any reason that this should not be implemented on LW as well.
OTOH, suicidal people often have very few options, so removing one is a big cost. Lots of reports suggests that suicide hotlines aren’t actually much better than random like-minded people, though it’s possible that callers underestimate how helpful they were. And contacting locals is among the most likely strategies to backfire.
I wasn’t suggesting banning users contemplating suicide. (Do any sites do that? Is it a good idea?)
Are these claims research-based?
I want research, too, but in lieu of that, I second “lots of reports”. Also, consider this reality:
Imagine you tell someone that you’re thinking about killing yourself. Say this person decided to call 911, and you end up admitted into the mental ward of a hospital.
Unless you have very good insurance (or something has changed), they’re likely to let you out in a week. I’ve heard of this happening to countless people. It is simply too expensive to keep them in there very long.
So, a week later, you’ve:
Missed a week worth of pay (from work).
Racked up tens of thousands of dollars in hospital bills (most likely your insurance won’t pay for it all.)
Missed a week of work. Some workplaces will find a reason to fire you after that. (Not all managers care that this is illegal.)
You had to live in a place full of crazy people every day for a week.
You had to sleep in a place full of crazy people, perhaps sharing a room with one, every night, for a week.
You’ve been treated like a crazy person for a week. (Barely allowed any possessions, unable to go anywhere, living apart from friends and family, having your life scheduled by the staff, etc.)
Your new prescription may not even kick in for several weeks!
Is your situation and mood:
A. Better
B. Worse
The last suicidal person I called 911 for was not provided any services, but was charged hundreds of dollars. Apparently, they stopped by, asked him a pile of questions, he managed to avoid being taken to the mental ward, and then they charged him out the ears for the visit.
He was quite upset with me. Quite. Moral of that story: Don’t IM about suicidal thoughts while the internet connection is flaking out and then say something like “I can’t go on” and then also fail to answer the now very concerned friend’s phone call right after.
It’s enough to make me think that someone should do a study to determine whether fatalities might increase after a visit to a mental hospital.
And as for the suicide hotlines—a lot of the operators are volunteers. I wouldn’t be surprised if the number of operators who are volunteers is like 90%+. I’m sure you have to be pretty special to be a suicide hotline volunteer (nerves of steel + screening) but I doubt that most of the operators are psychologists. This could probably be verified if you were willing to look up the largest suicide hotlines and see how they source their operators. For now, a quick example of operator sourcing:
Crisis Clinic Volunteering Page:
“Crisis Line phone workers come from all walks of life and educational backgrounds.” (All educational backgrounds means lots of people who are not licensed psychologists).
http://crisisclinic.org/get-involved/volunteer/#24hour
There is one study I’m aware of.
Ooh… click
It is interesting that the suicide rate for people who have just left a mental ward is over twice that of those who have just started anti-depressants, but:
A. Those who end up in the mental ward are more likely to be in that group because they actually attempted to kill themselves, or because they have a plan, whereas the anti-depressant patients and other groups of people might be experiencing suicidal ideation without an attempt or plan, or simply just depression without suicidal ideation.
B. People who plan or attempt to kill themselves are probably a lot more likely to attempt to kill themselves in the future than people who have never attempted and may not even be thinking about it.
In order to know whether admitting a suicidal person into a mental ward does more harm or more good, we’d need a study where they compare outpatient suicide rates with suicide rates in a control group. A good control group would need to consist of equally suicidal people who were not admitted into a mental ward. In order to ensure they were comparable, they’d need to track things like whether a suicide attempt was made, whether the person was on anti-depressants, etc. Also, if nobody had hospitalized the hypothetical control group, there’d be a reason for that. This reason would have to be the sort of thing that wouldn’t impact the suicide rate itself. For instance, if those people are living in a place with no legal way to hospitalize a person against their will, it might produce a great control group. On the other hand, if the control group mostly consisted of people who were not hospitalized because they are alone and have no family members or friends who care enough to call 911, this would not be a great control group.
Now I wonder if there are any places where you cannot legally be admitted to a mental ward against your will… or whether there might have been a point in the past when there was a place like this.
It’s hard to imagine that there would ever have been a place like this, but you never know.
You’re quite right that it’d be very hard to demonstrate a causal effect (and without having read the study itself—don’t have access -, I suspect the researchers didn’t even want to try).
Actually, I have no idea how that could be done in practice. For voluntary hospitalization, it would be helpful if one couldn’t be hospitalized against one’s will, but I’m not aware of a time and place where that would be the case, and don’t expect there to be any. So one can study that only outside the realm of “hospitalization-worthy” suicidality, by using patients who have been offered hospitalization, but declined it, as a control. My quick search turned up no indication of even that having been done.
And for forced hospitalization, it seems sort of impossible in principle to find a control group...
I also just found this, again I don’t have access, and unfortunately it doesn’t even have an abstract. Might be relevant, though, judging by the title.
Check for that first link on Pubmed. I was able to access the article that way.
I’m not finding this new article anywhere. The text in a Google book search where the study had been mentioned suggests that it’s probably not the type of study I’m looking for though.
I doubt this type of study exists. About the best we could do is to compare the suicide rate in an area with no legal method for hospitalizing suicidal people against their will (if that exists and the country is developed enough to keep such statistics reasonably well) with the suicide rate of a comparable area.
I looked at the results section of that abstract, but there’s an obvious reason why their correlation of an increase in suicides after hospitalization may not have Epiphany’s causal explanation.
I didn’t mean banning, just people thinking “They won’t help me, no point posting” or “Holy mackerel they’re going to call 911 on me and get me locked up” or “I would ask LW for help, but I can’t get over fretting about the reaction”.
A couple mental illness support groups I know of ban talking about planned suicides (and self-injury), mostly to avoid triggering potentially suicidal (resp. self-injurious) people. Since they point to emergency resources a lot and support non-acutely suicidal people, it doesn’t seem to be very bad, but I can’t observe the consequences.
The first is anecdotal. I’ve heard a lot of negative reports and very few positive ones, but there may be selection bias. Maybe hotlines are for people who are desperate to talk to anyone, so people who’d talk to me don’t benefit nearly as much.
The second… is probably wrong, come to think of it. A lot of people shriek “Never spill the beans about someone being suicidal without their consent, it could get them into arbitrary amounts of trouble, and an even suckier life won’t make them less suicidal”, but even more people shriek “Call 911 NAO” and the main evidence that the former know what they’re talking about is that they have much more personal experience.
Ooh. Ooh. I totally want these studies. If there is research. Is there? eager to get that research
While I agree in general, I do think such guidelines and agreements are best discussed in a quiet, separate thread, with explicit agreement of the concerned Mods, and not in response to an “emergency” call for help. In this instance, I am very glad that Slade did manage to call for help this way.
These are good points, Gwern, but the psychology industry and suicide hotlines fail a lot more often than you’d think (support for this point is included). If Slade has not tried them, they may still be useful for Slade. It is also possible that Slade has tried a few different psychologists, a few different prescriptions, and a few different hotlines without relief. Here are a few problems that one may run into when attempting to utilize the standard forms of help:
Note to Slade: You may not want to read this. This is for people who don’t have a clue why anybody would be asking here. It consists of a list of problems. It’s useful for the purpose it’s serving, but not particularly uplifting for a person in your situation. I’m writing you a separate response.
Psychologists failing to practice science. I’m sure you, Gwern, are aware how common this is but I’ll include a link for others.
Prescription drugs to treat moods not working. (See “A link for others”.)
Prescription drugs to treat moods causing intolerable side effects. (The FDA doesn’t approve drugs based on them being safe. It approves them based on whether the problem caused by the drugs is better or worse than the original affliction. I will use tardive dyskinesia, a neurological disorder resulting in involuntary body movements like repeatedly sticking one’s tongue out, as my example. Anti-psychotics can cause it.)
If Slade is gifted (not unlikely according to these numbers and Slade may have a 50% chance of not knowing (“about half of our country’s (America) gifted students are never identified”)) then they may find themselves being misdianosed with disorders they don’t have (happens to 25% of the gifted population (Citation: Misdiagnosis and Dual Diagnosis of Gifted Children and Adults) let alone the portion of the gifted population which actually shows up in a psychologist’s office).
If Slade is gifted, they might find that getting help from a psychologist who does not specialize in gifted clients is useless, frustrating or harmful. This can feel like taking a space ship to a car mechanic. Gifted people may need psychologists who have experience with gifted clients.
Suicide hotlines are non-profit organizations. As such they can end up understaffed. The following scenario does happen: Can you imagine how it would feel to call a suicide hotline and be told “Hold on a minute, I’m getting another call.” What if it kept happening over and over again throughout the call? If you look past the obvious annoyance that this is likely cause during your important conversation about this life or death matter, you’ll also see that you may begin to wonder how many other people are killing themselves while you receive the help. Total survivor’s guilt.
It is quite possible that Slade has attempted to get help using the stereotypical options and has run into one or all of these issues.
I don’t blame Slade for asking here. In the event that Slade’s main problem is that the psychologists don’t seem to get it—that Slade is gifted and they’re targeting their help to non-gifted clients, or that Slade is receiving unscientific advice from psychologists—one of the places where Slade would have a decent probability of finding these things out is by asking here.
If Slade has been failed by the ordinary methods, would you agree that the situation is important enough to “throw spaghetti at the wall” by posting on the internet asking for help, and that LessWrong would have a significantly better chance of producing a useful response than most websites?
Personally, I think the best approach when dealing with a problem this big is to talk to as many different people as possible about it—including both professionals and laymen that are either knowledgeable about the subject or generally smart. Nobody is infallible, and the more information you have, the better.
This sounds to me like taking the precautionary principle too far. Shouldn’t we do an expected value calculation? I imagine the folks at Wikipedia had a few incidents of people being helped with their depression and killing themselves subsequently (possibly as a result of the help they received), then Wikipedia was like “What should we do? How about a don’t-talk-with-suicidal-people policy?” without looking at the other, much more obscure cases where someone asked for help and did not subsequently commit suicide (possibly as a result of the help they received).
I seem to have read that talking to a random smart, empathetic person is just as good as talking to a pro counselor, BTW.
ETA: gwern, what’s your take on http://www.reddit.com/r/suicidewatch?
I don’t know. Wikipedia spends a lot less time on socializing then LW does.
No one can stop them even if it’s a terrible idea, they have nothing to lose, and nevertheless, they list their list of suicide hotlines prominently in the sidebar.
All fair points.
How much can you narrow down my identity from a comment, I really have no idea? That’d probably only cause trouble for me, by the way.
There’s really no reason someone couldn’t do those things even if they did have contributions under another account. And I’m curious, what do you mean by “attack” other than trolling?
There is a difference between a suicide note by someone determined to kill himself and a plea for advice from someone who doesn’t particularly want to die but might kill himself all the same. Your point about attempts to help plausibly backfiring still stands, though.
I don’t know; it depends on the details. I know from my time as an admin on Wikipedia that we were often able to contact local authorities or family, often using IP addresses. (I imagine suicide is quite troubling too.)
It reduces the odds.
I think that if you can’t think of ways in which suicides, real or faked, could be used to attack a group, then it’s probably better if I don’t list every idea that comes to mind.
I am not an expert on suicide (which is one reason I think a LW suicide policy is a really good idea), but I understand many or most suicide attempts fail so I’m not sure there is a difference or if the difference is important.