The New Dark Ages Is Long on Alarm, Short on Scientific Literacy
Reason · RC · trust 46/100
(Doubleday) The New Dark Ages: The Death of Reading and the Dawn of the Post-Literate Society , by James Marriott, Doubleday, 208 pages, $26
The last few years have seen a glut of sky-is-falling books about kids. Jonathan Haidt's The Anxious Generation has made a generation of parents anxious. Jared Horvath's The Digital Delusion has fueled a backlash against technology in schools. The latest, James Marriott's The New Dark Ages , is the inevitable end point of the genre: heavy on lofty moral lectures but stripped almost entirely of scientific literacy or curiosity.
The book's premise is straightforward: Reading and literacy have declined among younger generations, and this is leading not just to dumb, lazy kids, but to the end of democracy itself. The culprit: the smartphone.
On the positive side, The New Dark Ages has probably the best prose of its genre. The writing is earnest and engaging, with none of the clunkiness of the other books. Marriott's love for reading is infectious, and his concerns will be catnip for the audience of adults eager to hear, yet again, that kids today are dumber, sadder, and less prepared for democratic life than prior generations.
But its scientific literacy is even worse than the other books'. Haidt has been rightly criticized many times for cherry-picking evidence, overstating research conclusions, and producing simple graphs of the sort we warn students not to make causal inferences from in Psychology 101. Those graphs are now obsolete as well: The Centers for Disease Control and Prevention have documented that youth suicide , self-reported mental health problems , and self-injuries have all declined even as screen use remains at all-time highs. (That's one danger of using time-trend graphs to make causal assertions. Often the trends inconveniently reverse.) But The New Dark Ages mostly ignores scientific data altogether.
The book's references mention very few peer-reviewed studies, and the ones it does cite say little about the thesis that screens caused such changes. Instead, Marriott relies heavily on news reports, popular books, and similar secondary sources. This is unfortunate, as the data are clear that news media tend to sensationalize weak studies that suggest social media harms, while ignoring studies that suggest social media have basically no influence on youth mental health. He depends, for instance, on the Financial Times (he thanks one of its reporters for "talk[ing] me through the data") to suggest some trends, even though that outlet has been rightly criticized for using dodgy data in misleading and alarmist ways .
Like many alarmists, Marriott often presents data in ways that ignore a larger context. For instance, Marriott notes that America's standardized reading scores on the National Assessment of Educational Progress ( NAEP ) reached their lowest point in 20 years in 2024, with the implication that this is due to modern screens. He does not mention that NAEP scores actually peaked in the high-screen 2010s, declined after COVID, and now are back to levels seen in the early 1990s. Further, those post-COVID declines continued even after schools enforced more restrictive bans on cellphones and other technologies.
Further, the use of educational screens in schools is associated with slight improvements in reading and literacy, not declines (though those improvements are probably too small to be considered useful). There's a nuanced conversation to be had about screens and reading—much of which now happens on screens—but for a book that laments the loss of deep reading, Marriott doesn't seem to have bothered to do any deep reading in this area himself.
What's more, as researcher Mike Males recently pointed out , trend analysis suggests that the more kids use social media, the more they read books. I doubt very much that that's causal—but it's certainly a problem for the "screens are killing reading" hypothesis.
Another problem: Higher screen use is common for families that have a host of other troubles. Controlling for those other factors often makes the predictive value of screen time alone vanish.
Marriott presents various frightening-sounding anecdata about the decline in reading. Some of this is compelling—such as the overall decline in reading for pleasure, which began during the 1950s—and I don't want to discount every negative possibility here. But he presents all this as a catastrophic decline rather than a qualified change in society.
And I wish he were more skeptical about the common refrain that college students today can't read the books their professors assign. An entire cottage industry of such essays has sprung up, where professors ritualistically rend their clothes and gnash their teeth. But there are several potential problems with these analyses. For one thing, this concern isn't new. I've been teaching college for almost 30 years, and not once during that time have college professors thought their students were good readers. Also, some of the books assigned by professors may in fact be genuinely boring. Not everything on a course syllabus is a worthy classic.
Most important: If the average college student today is a poorer reader than 30 or 50 years ago, the reason may be more prosaic than anything related to television or smartphones. Universities are facing a demographic cliff, and many of them are being forced to recruit less-qualified students. The average IQ of college students is getting lower , not because young adults are stupider than in the past, but simply as a consequence of basic economics: Schools need warm bodies to keep the lights on.
If we're looking for patterns in time trends, we can't ignore the fact that young people today, on most outcomes, are far better than in earlier eras. Both youth violence and teen pregnancy are at record lows , more kids graduate high school than ever, and community volunteering is higher than in the 1970s and '80s. Smoking , drinking , and illegal drug use…
Read the original at Reason →
Open in TruthVane →