In a nutshell, Haidt believes:
Is this true? The book is fairly convincing in isolation—the spikes in suicide rates and ER visits for self-harm seem especially noteworthy1—but of course any position can seem compelling if you only hear from its proponents. A friend pointed me to the If Books Could Kill episode on this book, which throws a lot of cold water on Haidt’s representation of the data; this substack post collects links to several critical reviews, including one at Nature insisting:
Hundreds of researchers, myself [Candice Odgers, a professor at UC Irvine] included, have searched for the kind of large effects suggested by Haidt. Our efforts have produced a mix of no, small and mixed associations. Most data are correlative. When associations over time are found, they suggest not that social-media use predicts or causes depression, but that young people who already have mental-health problems use such platforms more often or in different ways from their healthy peers.2
Haidt acknowledges the existence of disagreement and seemingly-contradictory evidence on the issue (although I don’t think he gives any indication that he’s defending a minority or even fringe view within the academic community, which is the impression given by the podcast I linked above and perhaps the article I just quoted). In the text, he blames some confusion on research often lumping all technology and genders together, when the effect in question emerges only when looking specifically at social media’s effect on girls3, and some confusion on “look[ing] for effects of social media on individuals in isolation” instead of “group-level effects”4; in a footnote, he acknowledges ongoing disagreement, and refers readers to one of his substack posts.
I’m not going to try to get to the bottom of it; this review will just discuss some of my tangential takeaways.
In the past my instinctive reaction to worries about the effect of smartphones on kids has been basically: oh, c’mon 🙄. Earlier generations’ fears about new tech corrupting the youth were silly, and these contemporary fears are just more of the same, right?
No, that’s a glib reply, and I condemn past-Jacob’s intellectual laziness. First of all, obviously, the future isn’t always like the past, and actual evidence that something different is happening should trump mere intuition. Second, as mentioned in the If Books Could Kill episode I linked above, we may not even know (I personally certainly don’t know) that things actually did work out OK in the past. Maybe society is currently in the process of collapsing in part because some previous technological revolution systematically messed us all up in childhood in ways we can’t recognize.
Anyway, I think a major factor underlying the appeal of the glib reply, for me, is that I and most of my friends have been using smartphones and social media for a long time and it doesn’t seem like it’s been that big of a problem for us. But Haidt is arguing that there’s something specifically bad about being exposed to that tech in adolescence (which my generation was not—the goofy PocketPC/PalmPilot craze during my junior high years may have foreshadowed our screen obsession but was a far cry from the perpetual onlineness that smartphones encourage).
There seems to be a … sensitive period for cultural learning…. The Japanese anthropologist Yasuko Minoura studied the children of Japanese businessmen who had been transferred by their companies to live for a few years in California during the 1970s. She wanted to know at what age America shaped their sense of self, their feelings, and their ways of interacting with friends, even after they returned to Japan. The answer, she found, was between ages 9 and 14 or 15. Those children who spent a few years in California during that sensitive period came to “feel American.” If they returned to Japan at 15 or later, they had a harder time readjusting, or coming to “feel Japanese.” Those who didn’t arrive in America until age 15 had no such problems, because they never came to feel American, and those who returned to Japan well before 14 were able to readjust, because they were still in their sensitive period and could relearn Japanese ways. Minoura noted that “during the sensitive period, a cultural meaning system for interpersonal relationships appears to become a salient part of self-identity to which they are emotionally attached.”
So what happens to American children who generally get their first smartphone around the age of 11 and then get socialized into the cultures of Instagram, TikTok, video games, and online life for the rest of their teen years? … Any child who spends her sensitive period as a heavy user of social media will be shaped by the cultures of those sites. This may explain why Gen Z’s mental health outcomes are so much worse than those of the millennials: Gen Z was the first generation to go through puberty and the sensitive period for cultural learning on smartphones.5
Haidt lists “attention fragmentation” as one of “the four foundational harms of the new phone-based childhood”6, citing the study “Brain Drain: The Mere Presence of One’s Own Smartphone Reduces Available Cognitive Capacity”. Googling this brings up a 2023 meta-analysis which reaffirms the effect and even concludes:
Based on the available research findings, it seems advisable that smartphones should not even be near learners during periods of learning.7
(It sounds like this may have more to do with working memory than attention, but I haven’t dug in to the details.)
This is the issue that resonates most with me personally. I love the convenience, productivity, and safety benefits of phones, but I also often find myself thinking of the phone as an anchor: just knowing that it’s nearby keeps part of my consciousness tethered to the Internet. It’s hard to be fully focused on any activity for a prolonged period of time when I’m continuously aware that an (effectively) infinite number of other options are instantly accessible, many of which—like checking social media—offer a chance of a quick low-effort dopamine hit. But constant context-switching (or constant effort to resist the temptation of it) and endless scrolling are ultimately far less satisfying than losing myself fully in a task, and I’m increasingly resentful of feeling obligated to carry around a device that inhibits the latter. Stuffing the phone into my bag/backpack helps a little, but I feel like I don’t really start detoxing unless I get, like, hundreds of feet away from it.
One of Haidt’s proposals is for social media sites—along with porn sites—to actually enforce minimum age requirements. He’s quick to say that he doesn’t mean using your government ID card;8 he suggests other possibilities: “[u]sing a network of people to vouch for each other”, blockchain tokens, and biometrics.9 He’s not asking for everyone to standardize on a single approach; he “hope[s] that companies that want to enforce a minimum age will begin to offer a menu of options from which the user can pick.”10 He also suggests that outsourcing can address privacy concerns:
…have sites farm the job out to another company that simply reports back to the platform: yes or no. Old enough, or not old enough. If the age verification company was hacked, all that the world would learn is that the people in their database once had their age verified, not that they had used Pornhub or any other site.11
I don’t think that’s realistic. Not only would the company have to resist the financial pressure to collect and sell user data, it would also have to resist the pressure to build its product as cheaply as possible. Unless you invest substantial effort into ensuring that the system does not log sensitive information anywhere, there is a high risk of doing so completely by accident. Still, I think this section of the book did change my attitude toward the idea of age verification, from oh-god-no to skeptical-but-not-completely-opposed.
Haidt also floats the idea of having an operating-system-level, on-by-default “age check” feature that parents can enable or disable, which would cause the device to tell websites whether its user meets age requirements. I don’t think this would be as easy as the book implies, and there may be some downsides to think through (e.g., if a site is trying to exploit children, this would mean the child’s device is proactively advertising itself as part of a vulnerable group when the child visits the site). But it might be an idea worth exploring.
Haidt seems to think social media has a corrosive effect on society in general:
Social media is a fountain of bedevilments. It trains people to think in ways that are exactly contrary to the world’s wisdom traditions: Think about yourself first; be materialistic, judgmental, boastful, and petty; seek glory as quantified by likes and followers.12
…religious injunctions to be slower to judge and quicker to forgive are good for maintaining relationships and improving mental health. Social media trains people to do the opposite: Judge quickly and publicly, lest ye be judged for not judging whoever it is that we are all condemning today. Don’t forgive, or your team will attack you as a traitor.13
On the one hand, I do feel like at least some social media—Twitter especially, and Threads seems to be following suit—create an environment where the most highly prized character trait is a willingness to make immediate, confident, absolute, and preferably funny proclamations of moral condemnation. I worry a lot that this is making reconciliation or compromise more difficult and pushing us all toward some eventual catastrophic violent conflict.
On the other hand, you could have said many of the same things about cable news and talk radio. And the belief that we ought to show our commitment to our community’s values by denouncing outsiders in the strongest possible terms, without deigning to engage in conversation with them as equals, is far from new—it was a major frustration I had with fundamentalist Christianity.
Is society in general more judgmental and less forgiving now than it was 50 or 100 or 1000 years ago? I would be pretty shocked to hear a compelling argument that it is. What about now versus 5 or 10 years ago, though? Then I’m less sure.
Haidt also includes a Martin Luther King Jr. quote I hadn’t heard before but really like: “He who is devoid of the power to forgive is devoid of the power to love.” I find this insightful because it seems to be not just a noble-sounding platitude, but a true observation about reality—at least if you want to deeply love any actually-existing human for a prolonged period of time. Everyone makes some serious mistakes in their lifetime.
Jonathan Haidt, The anxious generation: how the great rewiring of childhood is causing an epidemic of mental illness (New York: Penguin Press, 2024), 30–31.↩︎
Candice L. Odgers, “The Great Rewiring: Is Social Media Really Behind an Epidemic of Teenage Mental Illness?” Nature 628, no. 8006 (March 29, 2024): 29–30, https://doi.org/10.1038/d41586-024-00902-2.↩︎
Haidt, The anxious generation, 146.↩︎
Ibid., 148–49.↩︎
Ibid., 63–64.↩︎
Ibid., 114.↩︎
Tobias Böttger, Michael Poschik, and Klaus Zierer, “Does the Brain Drain Effect Really Exist? A Meta-Analysis,” Behavioral Sciences 13, no. 9 (September 11, 2023): 7, https://doi.org/10.3390/bs13090751.↩︎
Haidt, The anxious generation, 236–37.↩︎
Ibid., 237.↩︎
Ibid., 237–38.↩︎
Ibid., 237.↩︎
Ibid., 209.↩︎
Ibid., 211.↩︎