Jonathan Haidt’s Book ‘The Anxious Generation’ Is Coddling The American Parent; Giving Them Clear, Simple & Wrong Explanations For What’s Ailing Teens

Jonathan Haidt’s new book, “The Anxious Generation,” has become a NY Times bestseller, and he’s making media appearances basically everywhere you look, telling people that social media has “rewired children’s minds” and that it is uniquely harmful.

We’ve talked about Haidt in the past, and especially his ability to consistently cherry-pick and misread the actual data on such things.

Haidt is telling a story a lot of people want to hear. The world has some very real problems happening these days, and it’s clear, simple, and ultimately wrong to be able to blame them on social media. It’s just that, in getting the diagnosis wrong, it lets people feel better about themselves, while brushing away actual real problems and doing the hard work of trying to dig up solutions.

The Daily Beast asked if I would review the book for them, and they’ve now published over 2000 words I wrote about the myriad problems of Haidt’s book. I’m not going to repeat it here, so please go read it over there, as I put a lot of time into that review.

But, as a quick summary: he’s wrong on the data, which undermines his entire argument. Almost every single expert in the field who does actual research on these issues says so. Candice Odgers ripped apart his misleading use of data in Nature. Andrew Przybylski, who has done multiple, detailed studies using massive amounts of data going back years, and keeps finding little to no evidence of the things Haidt claims, has talked about the problems in Haidt’s data. Ditto Jeff Hancock, at Stanford, who recently helped put together the National Academies of Sciences report on social media and adolescent health (which also did not find what Haidt found).

Indeed, one thing that came up in looking over the “strongest” research in the book was that (contrary to some of Haidt’s claims), data outside of the US on suicide rates seem to show they’re often (not always) going down, not up. Even worse, the data on depression in the US showing an increase in depression rates among kids is almost certainly due to changes in screening practices for depression and how suicide ideation is recorded.

As my review notes, though, the problems with the data are only the very beginning of the problems with the book. Because, in the first part of the book, Haidt misleadingly throws around all the data, but in the latter part, he focuses on his policy recommendations and basically comes up with a bunch of very silly ideas that have no data to back them up:

He suggests raising the age at which kids can use certain websites from 13 to 16. Why 16? Based on his gut. He literally says he “thinks” age 16 “was the right one for the minimum age,” but presents no research or data to explain why. He notes that at that age they’re mature enough to handle the internet, though he doesn’t explain why.

And why suggest limiting access to age 16, rather than teaching kids digital literacy and how to better use the internet to avoid harms? He doesn’t say. He just decides what he thinks is right.

Elsewhere, he argues that there’s really little downside to implementing his policy solutions, and the review tries to dig into just how wrong that is. Cutting off kids from methods of communication that many use to find their communities, or to communicate with far-flung friends and family, can be really harmful.

But, the thing that gets me the most is that anyone who has actually spent time on internet policy issues knows that every policy solution in the space involves very serious nuances and tradeoffs. Haidt jumps in to the deep end doing a YOLO belly flop with zero consideration of any of those tradeoffs.

He supports KOSA, despite the fact that pretty much everyone agrees it will do real harm to LGBTQ+ teens. He supports age verification, despite data protection experts noting that it’s a privacy nightmare and the Supreme Court saying it violates the First Amendment.

He tends to brush off any concerns in the manner of a person who is selling a book, but has never had to actually deal with the actual nuances, tradeoffs, and consequences of complicated policy decisions he doesn’t fully understand:

Some of Haidt’s suggestions are so disconnected from any actual research or data as to raise questions about exactly where he’s coming from. There’s an entire chapter talking about how the kids these days just need to be more spiritual and religious, which seems like an odd and out of place discussion in a book about social media (and, on a separate note there is at least some research suggesting that kids today are finding spirituality via social media).

When even his former co-author, Lukianoff, pointed out that Haidt’s proposals clearly violate the First Amendment, Haidt’s only response is to suggest that if First Amendment advocates get together, he’s sure they can figure out ways to do age verification that is Constitutional.

This is the classic “nerd harder” demands of a non-expert insisting that if actual experts try hard enough, surely they can make the impossible possible.

And my biggest concern in all this is that by playing up a new moral panic to sell books and the “Jonathan Haidt brand,” real harm is caused:

The actual harms to getting this wrong could be tremendous. By coddling the American parent, and letting them think they can cure what ails kids by simply limiting the internet access, real harm can be caused.

Kids who actually do rely on the internet to find community and social interactions could grow further isolated. Even worse, it stops parents and teachers from dealing with actual triggers and actual problems, allowing them to brush it off as “too much TikTok,” rather than whatever real cause might be at play. It also stops them from training kids how to use social media safely, which is an important skill these days.

Treating social media as inherently harmful for all kids (when the data, at best, suggests only a very small percentage struggle with it), also would remove a useful and helpful tool from many who can be taught to use it properly, to protect a small number of users who were not taught how to use it properly. Wouldn’t a better solution be to focus on helping everyone to use the tools properly and in an age appropriate manner?

As noted, there’s a lot more in there. Again, the full review clocks in at over 2000 words, but I’m hopeful that, even as Haidt’s book is getting widespread attention, people might, finally, begin to realize that he’s selling a bill of goods which appears to be a lot more harmful than the unproven harms he claims to be warning about.