Thanks for continuing to patiently respond to the skeptics with data-driven and logic-driven responses. I appreciate people want to hang on to long-held beliefs, but your work is important in helping come to the truth so we can make changes that will actually have positive impacts on our kids (or at least the next generation).
A major driver of mental health problems is changes in the omega-3/6 balance of the food supply due to increased arachidonic acid intake. But who is paying attention to that? For the latest word on that aspect of the suicide problem, read Omega Balance by Australian zoologist Anthony Hulbert, PhD.
Excerpt from a research paper: "The present findings suggest that, as naturally absorbed nutrients, higher EPA and lithium levels may be associated with less suicide attempt, and that higher arachidonic acid levels may be associated with more deliberate self-harm." (web search - Naturally absorbed polyunsaturated fatty acids, lithium, and suicide-related behaviors)
I was just thinking this morning how over-toxinned and undernourished our young people are. There definitely are physiological contributors to mental and emotional distress!!
James Greenblatt provides additional sources of info about such deficiencies. His company is “Psychiatry Redefined: We can do better.”
Well said. Physiologically speaking, molecular psychiatry would be more effective if, in addition to supplements, practitioners were to teach clients to reduce omea-6s intake.
Right!?! There are so many ways we can improve health by simply subtracting things — as you point out with omega 6 — rather than by just adding in more stuff.
Because that's where the studies are done. Define 'western civilation' and see if you can do it without referencing modern medical care and scientific research. Counter, if you will, with as many 3rd world studies on teen angst and social media addiction as you can find.
You fail to consider that perhaps it because these findings only occur in certain societies, which obviously leads to other possibilities. Perhaps all of the countries where the studies are done have other things in common with each other which all the other countries of the world do not. When your own civilization cannot answer a basic question such as "what is a woman", that is itself can be defined as mental illness, and we should not be surprised all of the variables that go into the alleged inability to answer that question (and the stark consequences for those who dare to give the "wrong" answer, which happens to be the common sense RIGHT answer), it should be no surprise that young people are confused, and that those already predisposed for mental illness are driven right into it.
> Perhaps all of the countries where the studies are done have other things in common with each other which all the other countries of the world do not.
Sure, so what might that be? Not all Western nations share a common culture, so what cross-cultural factors other than smartphones and social media exist? There probably aren't as many as you think.
That is exactly my point. There are other cultural variables present in the West that are not found in the rest of the world. These need to be explored as possible explanations for the problem on teen mental illness.
There is not only less stigma surrounding mental health issues, it's acually been glamorized. It's down to the victim mentality phenomenon.
I would argue that these things you blame on the internet and social media are more due to those platforms being captured by woke interests than by kids simply using them.
I agree that ‘destigmatization’ and ‘social contagion’ are in the same conceptual neighborhood. But to then go to ‘woke’ influence as the root cause is to argue on similar logic as those who say ‘of course, they’re depressed, the world--- racism, climate, economy--- has gone to hell since 2009’. The smartphone argument is that there is a psychological dynamic that is upstream of the language content. Which is that humans have always had a lot to worry about; worry is an adaptive safety mechanism. But historically humans find security in loyalty to and from a small group. The internet group dynamic has imbalanced this in an unhealthy way.
I understand the concept. I don't think it's the fault of the internet or social media, per se, though. A similar argument is that we could eliminate shooting deaths by eliminating guns. It's absolutely true, but it wouldn't necessarily eliminate murder or make the world a better place.
I think the gun analogy is pretty good. It’s been shown statistically that the immediate availability of a simple and effective way to commit suicide increases suicide risk. So, in this sense guns cause suicide. It’s in the same sense that the distorted internet dynamic causes psychological harm to adolescent girls. You can reasonably argue that the spot to intervene in the causal chain is at the depression for suicides and at the unhealthy self-image in girls.
But for me, the big question is whether we could ever have an evidence-informed public policy, which is what Haidt and Rausch are proposing in this case. The first step for that would be to get more clarity on the strong genetic preload of stereotypical human behaviors. To quote Iris Berent, “Our blindness to human nature is rooted in human nature itself.* With this approach we might at least avoid the harm done by the Blank-Slate utopian ideologies--- (such as the blindness of Marxism to the free-rider problem, for an off-topic example). This is why Haidt’s Moral Foundations approach to understanding human nature prompts him to look at the coalitional psychology at play upstream of the language content.
*The Blind Storyteller: How We Reason About Human Nature, by Iris Berent
Art imitating life and life imitating art. For a while now “media” has been producing content that ranges from books to tv to podcasts and YouTube and similar formats. Shows and discussions often present these “hard topics”, for instance in the 90’s I learned about bulemia (eating then induced vomiting to avoid weight gain); in the 00’s I learned about “cutting”; in the current age there seems to be unlimited variations where you hear about various mental health conditions and even the names of the psychiatric medications. It’s probably too much to say that the “art” is pushing some of these disorders as people self diagnose and everyone seems to have a therapist or has considered seeking out help. Adolescent children and those experiencing hormonal fluctuations associated with puberty and monthly cycles could be in a very vulnerable position. Isolation does not help, focusing on negative emotions and worries and ruminating on troubles, both real and perceived just solidifies the state of uncertainty and negative emotions.
Being outdoors, having a supportive peer group, friends, family members can be a positive influence and a good distraction when troubled intrusive thoughts prevail. There are disorders associated with inadequate sunlight exposure. Absence of father in the home, a chaotic and unstable family structure doesn’t help.
Into this situation that is primed for vulnerability we introduce social media and that is a sure recipe for disaster.
What you are doing is very important. Collecting the data, presenting it in a way that is easy to read and comprehend, this can lead to change. In time, hopefully sooner rather than later, your research and writing will lead to healthier outcomes for this population. It’s important to remember that the interests of corporations that are making money from tech, from psychiatric services and medication will all have their own theories based upon their biased perspective. Tech is good for some people, and harmful for others and teasing out the populations most at risk is a difficult task but you are doing an excellent job of it.
Never forget that the populations most at risk may be parented by the individuals least likely to access this material. I hope that your media interviews and lectures continue to grow and be widely available for everyone. Keep up the good work!
I find a lot of this persuasive. I had previously had the stigma objection and appreciate the way you address it.
However, I have a question about this point you make:
"As you saw in Figures 1 and 2 above, the rise in hospitalizations for self-harm began before 2015, around 2010 (long before the ICD change).
In fact, the average estimated yearly increase of adolescent girls (ages 10-19) admitted to hospitals from 2010 to 2015 (2,224 girls per year) was larger than it was from 2016 to 2020 (an additional 614 girls per year)."
My objection is that the rise in 2010 seems too early to say the phones were a driver of hospitalizations. Anecdotally, I got my first smartphone in 2012. At the time, it wasn't unusual to have a flip phone. I graduated from high school in 2009 and I knew one person who had an iPhone. Even for people who got a smartphone in 2009, wouldn't 2010 be too soon to see this sort of impact in the data? It seems unlikely to me that the effect on mental health was this immediate.
I don't think the authors do a good enough job of breaking down the causal factors here, although other articles break it down better. I think there are two independent factors, and they *reinforce* each other:
* smart phones with photos and video capability: being captured in photos or on video in any social setting can accentuate one's innate anxiety of being constantly watched, ridiculued or feeling free to act stupid in a way that won't cause long-term reputational damage.
* social media: makes distributing photos and videos easy, so even people who weren't present to understand the context could judge you.
Is it because girls are denied basic safety? Any man or teenager who declares herself a trans woman is allowed to freely enter women's locker rooms and women's restrooms. If a girl feels uncomfortable, she is accused of "transphobia," shamed and bullied.
But no one will dare to consider such a hypothesis because left-wing activists have intimidated everyone.
Did the same mental health decline happen in countries which are not trans-friendly?
You just call it "stupid" and claimed that "girls are not having mental health issues because they’re afraid of trans people in their locker rooms" without any proof. Stop it. It is silly and childish.
Teenage girls are at risk for suicide. A cause I am all too familiar with: the "Queen Bee" who runs the "popular" crowd that ruthlessly crushes any girl who does not meet their standards of beauty.
You have a tendency in this post (although not uniformly throughout) to argue against a strawman: Most of what I've seen is people arguing that differences in stigmatization and diagnosis could be a partial explanation for the trends your focused on. I haven't seen people arguing that those differences mean that cellphone usage couldn't be a factor.
Also, since you argue with such a high level of certainty about causation, you need to explain better why there is some inconsistency across countries with respect to these trends, why you're making such strong claims in the absence of clear data for many, many countries, why there have also been dramatic rises in mental health issues among other groups (like adult men) and how you've eliminated any potential causality in those cases as explanatory for teenage girls as well.
Finally, a problem I mentioned on Jonathan's previous post is that I think there's. big problem that came up in Jonathan's interview with Tyler Cowan, where it seems that Jonathan wasn't even sure that data on kids interacting with each other included time spent in schools. I get that there's a qualitative difference between time spent in unstructured activities and time spent in school, but it seems rather dubious to treat the impact of interaction time as dichotomous across those categories. Surely, time spent interacting in classroom, on school buses, on playgrounds, in lunchrooms, in intramural sports (not to mention in after school gymnastics classes and the like) have some measure of the benefits attributable also to free play time. As such, I don't get how you could conduct a thorough analysis if you don't look at school associated interaction time as a factor to adjust for when you are trying to measure the impact of the diminshment of free time outside of school.
A few responses to some of your points. Honestly, it would require a whole post to address them all well. But just a few quick notes.
I'll go from back to front.
1) I believe Jon was referencing the American Time Use Study. And they do not include time spent in school or work in that survey. That is clear in the methods.
2) That is a fair point to raise. This post was specifically about addressing the questions regarding the reality of the mental health crisis; not so much about the causal chain. There is of course, much more work to be done.
4) "You have a tendency in this post (although not uniformly throughout) to argue against a strawman"
I disagree with you here. First, I am *not* saying that all of it can be explained by our causal theory (I agree these trends can explain some of the rise). My broad point is that the destigmatization theory does not fit the data very well, nor does the changes in reporting. My *biggest* point is that there is a tendency to not take the self-harm data seriously because these measures are more imprecise than suicide. And I think that ends up being problematic if we are to understand what is happening with girls.
You might disagree with my reading, but that is very much how I see it playing out in this article:
Thanks for the reply, Zach. Engagement is always appreciated. I haven't read the book, I've read some of Jon's posts and seen some of his podcasts so please forgive my questions if they're based in ignorance of the work you've done.
With that caveat, then (given your clarification about the American Time Use Study) what about the implications of attributing much causality to the drop in interaction time, if kids still spend at least 6 hours (often more) interacting in school or school-related activity or other activity like sports or band, etc.?
Yes, interaction during non-school time could reasonably be viewed as having dropped markedly (although a lot of kids have for years spend many hours watching TV or other less social activity), and perhaps that's more true for girls than boys (who have long spent much time playing video games). However, say (for the sake of argument) that time was generally 4 hours a day (during weekdays) and it's dropped by 75%, that's a limited change if there remains 6-8 hours a day of interaction time in school. Let's add in a degradation of quality of interaction time in school and school related activity because of cellphones, but certainly it's not a 100% degradation, or I'd argue anything close to that.
My point being, a significant percentage of the <i>total</i> previous interaction time has effectively remained - which I would think needs to be reconciled with your attribution of a huge effect from the non-school time drop in interaction time.
To that I'd add in what Cowan raised in his interview - that the enormous drop in interaction time during the pandemic school closures didn't show as a signal in your data (taking his word on that, I haven't looked but Jon didn't object to that assertion by Cowan).
Then you'd have to adjust for the direct impact of loss of time spent in moderate- to vigorous physical activity, which could have a direct impact on mental health independently of the impact of time spent on social media per say.
Just seems like all of this is very noisy for such confident assertions of a clear causality. Of course your basic thesis seems to have a strong plausibility.
------
As for the destigmatization/reporting aspect...I understand that they are of a questionable fit with the data and that they aren't reasons to dismiss the self-harm implications. There seem valid reasons to consider self-harm and suicide as both important and distinct independent outcome measures. But I did read that Vox piece and I didn't get from it a reading like yours. I saw it as recognizing that there are reasonable reasons to think that the underlying thesis deserve merit despite the questions about the metrics
Thanks for this Joshua. I really appreciate how deep you are going into this.
Back to front again:
1. That's good; maybe my reading is a bit biased. It is lines like this though that make me frustrated:
"In view of these measurement distortions and the falling youth suicide rates in Europe, Haidt’s critics argue that it’s unclear whether there even is a deepening teen mental health crisis that requires explanation."
This is where it goes from there are some imprecisions to there is nothing going on.
2. These are all excellent point. I really think to do justice to your questions would be to read chapters 2-7 in the book. It really lays out the broader scoping argument around the loss of the play-based childhood and the rise of the phone-based childhood. It is not only about time displacement, there are many other mechanisms of harm and how we believe it transformed social interactions. I think it would be difficult to fully address the question here. And if you want, I'd be happy to correspond by email after reading that part of the book, to continue the discussion
Thanks again, Zach. I will also own up to bias myself. But I read in that article where Eric kept saying there's reasonable reason to see BOTH sides as overstating the case. I suspect you're reacting disproportionately to his criticism of your "side" in a way that doesn't incorporate his criticism of the other "side"
Not sure if I'll read the book, but if I do I'll follow up if my questions about time use that remain. I do get that it's not only time displacement, and I get that there would be a quality reduction in critical measures comparing open, play-based childhood and time spend in school and school-related activities. I don't question any of that.
So then there are some separate questions regarding to what degree it's a loss of play-based time versus time replaced more specifically by social media. IOW, people have been talking about the impact of "helicopter parenting" for quite a while and disaggregating that from the impact of "social media" to identify a distinct causality seems to me to be extremely complex. So I'm reflexively skeptical more broadly, and I'm skeptical that the book would really address these questions and I prefer a direct interaction as a way to interrogate my own questions. But I'm sure you've got a lot of demands on your time and I have to be realistic. And I do appreciate the time you did take.
The adults are even less “alright.” If I were dictator, the following disclaimers would head all commentaries (citing ages of regular posters on here as examples) deploring teen mental health, suicide, self-harm, and/or social media:
- This humble respondent, male, age 73, belongs to a demographic Centers for Disease Control figures show is 6.3 times more likely to commit suicide and 4.7 times more likely to die from self-inflicted suicide or overdose than a high school girl (age 17).
- A popular author on here, male, age 60, belongs to a demographic 6.4 times more likely to commit suicide and 12.6 times more likely to die from self-inflicted suicide or overdose than a high-school girl. (Why are middle-aged men, including many fathers, so appallingly self-destructive and getting worse?)
- Another popular author, female, age 52, belongs to a demographic 1.7 times more likely to commit suicide and 5.1 times more likely to die from self-inflicted suicide or overdose than a high-school girl.
- “Not-alright” adults are by far the biggest cause of “not-alright” teens. Parents’ abuse, mental health, and rapidly rising drug/alcohol crises across the Anglo world are by far the leading known cause of teens’ mental health problems. No other factor is even close.
- The Centers for Disease Control’s definitive, multi-factor 2021 survey shows girls who are violently and/or emotionally abused by parents or household grownups are 8 times more likely to attempt suicide and 27 times more likely to self-harm compared to girls who are not abused. Why do none of the popular authors mention this?
- Social media use has mixed, generally protective associations. The CDC finds girls who frequently (5+ hours a day) use screens like TV, cellphones, and social media are 23% LESS likely to attempt suicide, 51% less likely to self-harm, and generally less likely to suffer other serious risks than are girls with little or no screen time (<1 hour a day). Again, why is this not mentioned?
All of the above statements are true and easily verified. Ordinarily, we would expect social scientists, political and health leaders, and media reporters to provide vital, balanced contexts when they discuss important issues like suicide and self-harm. We would also expect humility. Teenage girls are coping with their supposed stresses, including more messed-up adults, far better than are middle-agers, especially men.
There was an abrupt, large rise in self-harm and suicide rates for adolescent women in the 5 Anglophone countries starting around 2011. The cause of this change in rate is what Zach is examining. Are you saying that parent or household grownup violence and/or emotional abuse of girls has increased across the Anglosphere starting around 2011? Adults have become more hostile to children in the past decade?
Well-said as usual, Mike. I would also add the following advice to the fearmongering commentators: "Be a mentor, not a tormentor". As well as the famous quote by Gandhi, "Be the change you wish to see in the world".
One further question is the self-reinforcement of the whole psych industry.
Girls have, eg, been clearly demonstrated to have susceptibility to social fads than have boys - body image in particular (but not exclusively) anorexia/bullemia, the whole transgender child-abuse scam, fashion victimhood…you name it, really (hoola hoops, hair bobbing…). The question is about the influence of “grooming”
Anyone who works with children can valid all of what you are saying. The decline is rapid, visible and unmistakable. And, I don't believe we are at the bottom of the curve, not by a long way.
Playing devil's advocate - on challenge 1 - could social media have caused de-stigmatization, which has then caused more young people (girls especially) presenting with self harm injuries? I get that hospitalization is probably a high bar to clear, but when you look at the definition of self harm at the CDC it could refer to some relatively minor injuries. And if you've got a young woman in your hospital with self harm injuries, even if they're not that major you might still want to keep them in for monitoring (I've never worked in a hospital so no idea how likely that is).
Aside from that I think this is a really excellent dive into the whole 'is it just reporting/everyone being more comfy talking about mental health' part of this debate. The Economist graph did it for me. Because it is deaths by suicide (which is unlikely down to reporting changes and cannot be due to level of comfort talking about mental health) and clearly shows the impact is on young females only. It shows it isn't confined to one country. It hits the nail on the head in terms of setting out the problem, and is hard to refute (assuming the data analysis is good of course!).
One note (and I don't have a solution!) - although these long articles are great for people like me (and the rest of us subscribers) who have at least some attention span, you'd lose most people before you get past the first couple of paragraphs. You guys must be asked onto the TV and podcasts and stuff all the time - do you have simple, effective comms to get all this detail distilled down? A bit like the Economist graph (even that could do with some simplifying to be honest) - getting a tonne of information across in a small bite sized chunk. It feels like it should be kicking up a stink with lawmakers but I just don't see it being addressed hard enough.
"We think it is time to start taking action." I am very sympathetic to your goals and your agenda on After Babel. But when I read exhortations like this by those who want to turn back the tide of Western Liberalism's headlong self-destruct I have to say they make me wince a bit. Why ? Because these exhortations betray a somewhat romantic and naive misunderstanding of where we currently are in 2024. "We" can make noise Yes and that's what you are doing and all power to your elbow. But in our intrinsically out-of-control 21st c. culture "we need to take action" is - sadly - fairly meaningless.
Thanks for continuing to patiently respond to the skeptics with data-driven and logic-driven responses. I appreciate people want to hang on to long-held beliefs, but your work is important in helping come to the truth so we can make changes that will actually have positive impacts on our kids (or at least the next generation).
A major driver of mental health problems is changes in the omega-3/6 balance of the food supply due to increased arachidonic acid intake. But who is paying attention to that? For the latest word on that aspect of the suicide problem, read Omega Balance by Australian zoologist Anthony Hulbert, PhD.
Excerpt from a research paper: "The present findings suggest that, as naturally absorbed nutrients, higher EPA and lithium levels may be associated with less suicide attempt, and that higher arachidonic acid levels may be associated with more deliberate self-harm." (web search - Naturally absorbed polyunsaturated fatty acids, lithium, and suicide-related behaviors)
I was just thinking this morning how over-toxinned and undernourished our young people are. There definitely are physiological contributors to mental and emotional distress!!
James Greenblatt provides additional sources of info about such deficiencies. His company is “Psychiatry Redefined: We can do better.”
I love that tagline. So true.
Well said. Physiologically speaking, molecular psychiatry would be more effective if, in addition to supplements, practitioners were to teach clients to reduce omea-6s intake.
Right!?! There are so many ways we can improve health by simply subtracting things — as you point out with omega 6 — rather than by just adding in more stuff.
The world is more than just the West. Why isn't this increase in teen mental illness also found in the rest of the world?
How do you know it isn't?
Why do you think these studies only include English speaking countries/Western civilization?
Because that's where the studies are done. Define 'western civilation' and see if you can do it without referencing modern medical care and scientific research. Counter, if you will, with as many 3rd world studies on teen angst and social media addiction as you can find.
When Mahatma Gandhi was asked what he thought of Western Civilization, he replied, "I think it would be a great idea". Zing!
You fail to consider that perhaps it because these findings only occur in certain societies, which obviously leads to other possibilities. Perhaps all of the countries where the studies are done have other things in common with each other which all the other countries of the world do not. When your own civilization cannot answer a basic question such as "what is a woman", that is itself can be defined as mental illness, and we should not be surprised all of the variables that go into the alleged inability to answer that question (and the stark consequences for those who dare to give the "wrong" answer, which happens to be the common sense RIGHT answer), it should be no surprise that young people are confused, and that those already predisposed for mental illness are driven right into it.
Still, you can't find what no one is looking for.
> Perhaps all of the countries where the studies are done have other things in common with each other which all the other countries of the world do not.
Sure, so what might that be? Not all Western nations share a common culture, so what cross-cultural factors other than smartphones and social media exist? There probably aren't as many as you think.
That is exactly my point. There are other cultural variables present in the West that are not found in the rest of the world. These need to be explored as possible explanations for the problem on teen mental illness.
There is not only less stigma surrounding mental health issues, it's acually been glamorized. It's down to the victim mentality phenomenon.
I would argue that these things you blame on the internet and social media are more due to those platforms being captured by woke interests than by kids simply using them.
I agree that ‘destigmatization’ and ‘social contagion’ are in the same conceptual neighborhood. But to then go to ‘woke’ influence as the root cause is to argue on similar logic as those who say ‘of course, they’re depressed, the world--- racism, climate, economy--- has gone to hell since 2009’. The smartphone argument is that there is a psychological dynamic that is upstream of the language content. Which is that humans have always had a lot to worry about; worry is an adaptive safety mechanism. But historically humans find security in loyalty to and from a small group. The internet group dynamic has imbalanced this in an unhealthy way.
I understand the concept. I don't think it's the fault of the internet or social media, per se, though. A similar argument is that we could eliminate shooting deaths by eliminating guns. It's absolutely true, but it wouldn't necessarily eliminate murder or make the world a better place.
I think the gun analogy is pretty good. It’s been shown statistically that the immediate availability of a simple and effective way to commit suicide increases suicide risk. So, in this sense guns cause suicide. It’s in the same sense that the distorted internet dynamic causes psychological harm to adolescent girls. You can reasonably argue that the spot to intervene in the causal chain is at the depression for suicides and at the unhealthy self-image in girls.
But for me, the big question is whether we could ever have an evidence-informed public policy, which is what Haidt and Rausch are proposing in this case. The first step for that would be to get more clarity on the strong genetic preload of stereotypical human behaviors. To quote Iris Berent, “Our blindness to human nature is rooted in human nature itself.* With this approach we might at least avoid the harm done by the Blank-Slate utopian ideologies--- (such as the blindness of Marxism to the free-rider problem, for an off-topic example). This is why Haidt’s Moral Foundations approach to understanding human nature prompts him to look at the coalitional psychology at play upstream of the language content.
*The Blind Storyteller: How We Reason About Human Nature, by Iris Berent
Art imitating life and life imitating art. For a while now “media” has been producing content that ranges from books to tv to podcasts and YouTube and similar formats. Shows and discussions often present these “hard topics”, for instance in the 90’s I learned about bulemia (eating then induced vomiting to avoid weight gain); in the 00’s I learned about “cutting”; in the current age there seems to be unlimited variations where you hear about various mental health conditions and even the names of the psychiatric medications. It’s probably too much to say that the “art” is pushing some of these disorders as people self diagnose and everyone seems to have a therapist or has considered seeking out help. Adolescent children and those experiencing hormonal fluctuations associated with puberty and monthly cycles could be in a very vulnerable position. Isolation does not help, focusing on negative emotions and worries and ruminating on troubles, both real and perceived just solidifies the state of uncertainty and negative emotions.
Being outdoors, having a supportive peer group, friends, family members can be a positive influence and a good distraction when troubled intrusive thoughts prevail. There are disorders associated with inadequate sunlight exposure. Absence of father in the home, a chaotic and unstable family structure doesn’t help.
Into this situation that is primed for vulnerability we introduce social media and that is a sure recipe for disaster.
What you are doing is very important. Collecting the data, presenting it in a way that is easy to read and comprehend, this can lead to change. In time, hopefully sooner rather than later, your research and writing will lead to healthier outcomes for this population. It’s important to remember that the interests of corporations that are making money from tech, from psychiatric services and medication will all have their own theories based upon their biased perspective. Tech is good for some people, and harmful for others and teasing out the populations most at risk is a difficult task but you are doing an excellent job of it.
Never forget that the populations most at risk may be parented by the individuals least likely to access this material. I hope that your media interviews and lectures continue to grow and be widely available for everyone. Keep up the good work!
I find a lot of this persuasive. I had previously had the stigma objection and appreciate the way you address it.
However, I have a question about this point you make:
"As you saw in Figures 1 and 2 above, the rise in hospitalizations for self-harm began before 2015, around 2010 (long before the ICD change).
In fact, the average estimated yearly increase of adolescent girls (ages 10-19) admitted to hospitals from 2010 to 2015 (2,224 girls per year) was larger than it was from 2016 to 2020 (an additional 614 girls per year)."
My objection is that the rise in 2010 seems too early to say the phones were a driver of hospitalizations. Anecdotally, I got my first smartphone in 2012. At the time, it wasn't unusual to have a flip phone. I graduated from high school in 2009 and I knew one person who had an iPhone. Even for people who got a smartphone in 2009, wouldn't 2010 be too soon to see this sort of impact in the data? It seems unlikely to me that the effect on mental health was this immediate.
I don't think the authors do a good enough job of breaking down the causal factors here, although other articles break it down better. I think there are two independent factors, and they *reinforce* each other:
* smart phones with photos and video capability: being captured in photos or on video in any social setting can accentuate one's innate anxiety of being constantly watched, ridiculued or feeling free to act stupid in a way that won't cause long-term reputational damage.
* social media: makes distributing photos and videos easy, so even people who weren't present to understand the context could judge you.
Am I misreading the chart? It seems like teen girls are by far the least likely demographic to commit suicide.
Certainly. Author himself acknowledges that the point is the rise in suicide rate which is opposite to drop in suicide among most other groups.
Is it because girls are denied basic safety? Any man or teenager who declares herself a trans woman is allowed to freely enter women's locker rooms and women's restrooms. If a girl feels uncomfortable, she is accused of "transphobia," shamed and bullied.
But no one will dare to consider such a hypothesis because left-wing activists have intimidated everyone.
Did the same mental health decline happen in countries which are not trans-friendly?
You just call it "stupid" and claimed that "girls are not having mental health issues because they’re afraid of trans people in their locker rooms" without any proof. Stop it. It is silly and childish.
"Exaggerated" is the proper spelling.
However, this is a great article.
Teenage girls are at risk for suicide. A cause I am all too familiar with: the "Queen Bee" who runs the "popular" crowd that ruthlessly crushes any girl who does not meet their standards of beauty.
Oh gosh. I can't believe I missed that. Thanks.
I've been a journalist for 44 years and did a LOT of copy-editing.
Headlines are the toughest things...the point size is so big, you can't believe you misspelled the word.
You have a tendency in this post (although not uniformly throughout) to argue against a strawman: Most of what I've seen is people arguing that differences in stigmatization and diagnosis could be a partial explanation for the trends your focused on. I haven't seen people arguing that those differences mean that cellphone usage couldn't be a factor.
Also, since you argue with such a high level of certainty about causation, you need to explain better why there is some inconsistency across countries with respect to these trends, why you're making such strong claims in the absence of clear data for many, many countries, why there have also been dramatic rises in mental health issues among other groups (like adult men) and how you've eliminated any potential causality in those cases as explanatory for teenage girls as well.
Finally, a problem I mentioned on Jonathan's previous post is that I think there's. big problem that came up in Jonathan's interview with Tyler Cowan, where it seems that Jonathan wasn't even sure that data on kids interacting with each other included time spent in schools. I get that there's a qualitative difference between time spent in unstructured activities and time spent in school, but it seems rather dubious to treat the impact of interaction time as dichotomous across those categories. Surely, time spent interacting in classroom, on school buses, on playgrounds, in lunchrooms, in intramural sports (not to mention in after school gymnastics classes and the like) have some measure of the benefits attributable also to free play time. As such, I don't get how you could conduct a thorough analysis if you don't look at school associated interaction time as a factor to adjust for when you are trying to measure the impact of the diminshment of free time outside of school.
Hi Joshua,
Thanks for this.
A few responses to some of your points. Honestly, it would require a whole post to address them all well. But just a few quick notes.
I'll go from back to front.
1) I believe Jon was referencing the American Time Use Study. And they do not include time spent in school or work in that survey. That is clear in the methods.
2) That is a fair point to raise. This post was specifically about addressing the questions regarding the reality of the mental health crisis; not so much about the causal chain. There is of course, much more work to be done.
Have you read Jon's response to skeptics:
1) https://www.afterbabel.com/p/phone-based-childhood-cause-epidemic
2) https://www.afterbabel.com/p/why-some-researchers-think-im-wrong
3) Ch. 2 through 7 in the book?
We try to make the case as strongly as possible.
And my work on European variation trends:
https://www.afterbabel.com/p/international-crisis-europe
4) "You have a tendency in this post (although not uniformly throughout) to argue against a strawman"
I disagree with you here. First, I am *not* saying that all of it can be explained by our causal theory (I agree these trends can explain some of the rise). My broad point is that the destigmatization theory does not fit the data very well, nor does the changes in reporting. My *biggest* point is that there is a tendency to not take the self-harm data seriously because these measures are more imprecise than suicide. And I think that ends up being problematic if we are to understand what is happening with girls.
You might disagree with my reading, but that is very much how I see it playing out in this article:
https://www.vox.com/24127431/smartphones-young-kids-children-parenting-social-media-teen-mental-health
And I am planning to write a post addressing the "intergenerational hypothesis."
According to the Vox article, the truth really seems to be somewhere in the middle.
Thanks for the reply, Zach. Engagement is always appreciated. I haven't read the book, I've read some of Jon's posts and seen some of his podcasts so please forgive my questions if they're based in ignorance of the work you've done.
With that caveat, then (given your clarification about the American Time Use Study) what about the implications of attributing much causality to the drop in interaction time, if kids still spend at least 6 hours (often more) interacting in school or school-related activity or other activity like sports or band, etc.?
Yes, interaction during non-school time could reasonably be viewed as having dropped markedly (although a lot of kids have for years spend many hours watching TV or other less social activity), and perhaps that's more true for girls than boys (who have long spent much time playing video games). However, say (for the sake of argument) that time was generally 4 hours a day (during weekdays) and it's dropped by 75%, that's a limited change if there remains 6-8 hours a day of interaction time in school. Let's add in a degradation of quality of interaction time in school and school related activity because of cellphones, but certainly it's not a 100% degradation, or I'd argue anything close to that.
My point being, a significant percentage of the <i>total</i> previous interaction time has effectively remained - which I would think needs to be reconciled with your attribution of a huge effect from the non-school time drop in interaction time.
To that I'd add in what Cowan raised in his interview - that the enormous drop in interaction time during the pandemic school closures didn't show as a signal in your data (taking his word on that, I haven't looked but Jon didn't object to that assertion by Cowan).
Then you'd have to adjust for the direct impact of loss of time spent in moderate- to vigorous physical activity, which could have a direct impact on mental health independently of the impact of time spent on social media per say.
Just seems like all of this is very noisy for such confident assertions of a clear causality. Of course your basic thesis seems to have a strong plausibility.
------
As for the destigmatization/reporting aspect...I understand that they are of a questionable fit with the data and that they aren't reasons to dismiss the self-harm implications. There seem valid reasons to consider self-harm and suicide as both important and distinct independent outcome measures. But I did read that Vox piece and I didn't get from it a reading like yours. I saw it as recognizing that there are reasonable reasons to think that the underlying thesis deserve merit despite the questions about the metrics
Thanks for this Joshua. I really appreciate how deep you are going into this.
Back to front again:
1. That's good; maybe my reading is a bit biased. It is lines like this though that make me frustrated:
"In view of these measurement distortions and the falling youth suicide rates in Europe, Haidt’s critics argue that it’s unclear whether there even is a deepening teen mental health crisis that requires explanation."
This is where it goes from there are some imprecisions to there is nothing going on.
2. These are all excellent point. I really think to do justice to your questions would be to read chapters 2-7 in the book. It really lays out the broader scoping argument around the loss of the play-based childhood and the rise of the phone-based childhood. It is not only about time displacement, there are many other mechanisms of harm and how we believe it transformed social interactions. I think it would be difficult to fully address the question here. And if you want, I'd be happy to correspond by email after reading that part of the book, to continue the discussion
Thanks again, Zach. I will also own up to bias myself. But I read in that article where Eric kept saying there's reasonable reason to see BOTH sides as overstating the case. I suspect you're reacting disproportionately to his criticism of your "side" in a way that doesn't incorporate his criticism of the other "side"
Not sure if I'll read the book, but if I do I'll follow up if my questions about time use that remain. I do get that it's not only time displacement, and I get that there would be a quality reduction in critical measures comparing open, play-based childhood and time spend in school and school-related activities. I don't question any of that.
So then there are some separate questions regarding to what degree it's a loss of play-based time versus time replaced more specifically by social media. IOW, people have been talking about the impact of "helicopter parenting" for quite a while and disaggregating that from the impact of "social media" to identify a distinct causality seems to me to be extremely complex. So I'm reflexively skeptical more broadly, and I'm skeptical that the book would really address these questions and I prefer a direct interaction as a way to interrogate my own questions. But I'm sure you've got a lot of demands on your time and I have to be realistic. And I do appreciate the time you did take.
Oh, yes. I have no issues with Eric. My concern is with the perspective of the researchers making the claim that there is nothing there.
I hope you will read the book!
The adults are even less “alright.” If I were dictator, the following disclaimers would head all commentaries (citing ages of regular posters on here as examples) deploring teen mental health, suicide, self-harm, and/or social media:
- This humble respondent, male, age 73, belongs to a demographic Centers for Disease Control figures show is 6.3 times more likely to commit suicide and 4.7 times more likely to die from self-inflicted suicide or overdose than a high school girl (age 17).
- A popular author on here, male, age 60, belongs to a demographic 6.4 times more likely to commit suicide and 12.6 times more likely to die from self-inflicted suicide or overdose than a high-school girl. (Why are middle-aged men, including many fathers, so appallingly self-destructive and getting worse?)
- Another popular author, female, age 52, belongs to a demographic 1.7 times more likely to commit suicide and 5.1 times more likely to die from self-inflicted suicide or overdose than a high-school girl.
- “Not-alright” adults are by far the biggest cause of “not-alright” teens. Parents’ abuse, mental health, and rapidly rising drug/alcohol crises across the Anglo world are by far the leading known cause of teens’ mental health problems. No other factor is even close.
- The Centers for Disease Control’s definitive, multi-factor 2021 survey shows girls who are violently and/or emotionally abused by parents or household grownups are 8 times more likely to attempt suicide and 27 times more likely to self-harm compared to girls who are not abused. Why do none of the popular authors mention this?
- Social media use has mixed, generally protective associations. The CDC finds girls who frequently (5+ hours a day) use screens like TV, cellphones, and social media are 23% LESS likely to attempt suicide, 51% less likely to self-harm, and generally less likely to suffer other serious risks than are girls with little or no screen time (<1 hour a day). Again, why is this not mentioned?
All of the above statements are true and easily verified. Ordinarily, we would expect social scientists, political and health leaders, and media reporters to provide vital, balanced contexts when they discuss important issues like suicide and self-harm. We would also expect humility. Teenage girls are coping with their supposed stresses, including more messed-up adults, far better than are middle-agers, especially men.
Bob Dylan said it best, I think:
"Come mothers and fathers
Throughout the land
And don't criticize
What you can't understand
Your sons and your daughters
Are beyond your command
Your old road is rapidly agin'
Please get out of the new one
If you can't lend your hand
For the times they are a-changin'"
There was an abrupt, large rise in self-harm and suicide rates for adolescent women in the 5 Anglophone countries starting around 2011. The cause of this change in rate is what Zach is examining. Are you saying that parent or household grownup violence and/or emotional abuse of girls has increased across the Anglosphere starting around 2011? Adults have become more hostile to children in the past decade?
Well-said as usual, Mike. I would also add the following advice to the fearmongering commentators: "Be a mentor, not a tormentor". As well as the famous quote by Gandhi, "Be the change you wish to see in the world".
One further question is the self-reinforcement of the whole psych industry.
Girls have, eg, been clearly demonstrated to have susceptibility to social fads than have boys - body image in particular (but not exclusively) anorexia/bullemia, the whole transgender child-abuse scam, fashion victimhood…you name it, really (hoola hoops, hair bobbing…). The question is about the influence of “grooming”
Anyone who works with children can valid all of what you are saying. The decline is rapid, visible and unmistakable. And, I don't believe we are at the bottom of the curve, not by a long way.
Playing devil's advocate - on challenge 1 - could social media have caused de-stigmatization, which has then caused more young people (girls especially) presenting with self harm injuries? I get that hospitalization is probably a high bar to clear, but when you look at the definition of self harm at the CDC it could refer to some relatively minor injuries. And if you've got a young woman in your hospital with self harm injuries, even if they're not that major you might still want to keep them in for monitoring (I've never worked in a hospital so no idea how likely that is).
Aside from that I think this is a really excellent dive into the whole 'is it just reporting/everyone being more comfy talking about mental health' part of this debate. The Economist graph did it for me. Because it is deaths by suicide (which is unlikely down to reporting changes and cannot be due to level of comfort talking about mental health) and clearly shows the impact is on young females only. It shows it isn't confined to one country. It hits the nail on the head in terms of setting out the problem, and is hard to refute (assuming the data analysis is good of course!).
One note (and I don't have a solution!) - although these long articles are great for people like me (and the rest of us subscribers) who have at least some attention span, you'd lose most people before you get past the first couple of paragraphs. You guys must be asked onto the TV and podcasts and stuff all the time - do you have simple, effective comms to get all this detail distilled down? A bit like the Economist graph (even that could do with some simplifying to be honest) - getting a tonne of information across in a small bite sized chunk. It feels like it should be kicking up a stink with lawmakers but I just don't see it being addressed hard enough.
"We think it is time to start taking action." I am very sympathetic to your goals and your agenda on After Babel. But when I read exhortations like this by those who want to turn back the tide of Western Liberalism's headlong self-destruct I have to say they make me wince a bit. Why ? Because these exhortations betray a somewhat romantic and naive misunderstanding of where we currently are in 2024. "We" can make noise Yes and that's what you are doing and all power to your elbow. But in our intrinsically out-of-control 21st c. culture "we need to take action" is - sadly - fairly meaningless.
shhhhhh!
dont tell liberals that the postliberal future has arrived.
you don't tell small children there's no such thing as Santa, do you?
Any study of how porn consumption and porn bulling ( and porn misogyny) by boys affects girls? \\