The Students Who Can’t Finish a Book
Professors are quietly shrinking their assigned reading syllabi. The data says the problem is real. The collective response says everything about what college thinks it’s for.
Last week I argued the old college model is dead. This is what its death looks like from inside a seminar room.
In 2024, The Atlantic published a piece whose headline did half the work: “The Elite College Students Who Can’t Read Books.” Rose Horowitch interviewed 33 professors, and their stories rhymed. Nicholas Dames, who has taught Literature Humanities at Columbia for decades, described students arriving “bewildered” by the expectation of reading multiple books in a semester. Others described the quiet workaround already underway: assign excerpts instead of books, chapters instead of novels, shrink the syllabus until it fits the students you have rather than the students you remember.
Let me do to this story what I’d ask you to do to any story: read it deeply before accepting my argument.
The honest caveat first. Horowitch’s evidence is anecdotal; 33 professors relaying impressions, not a dataset. Her own piece concedes that no comprehensive data exist on the trend, and critics said so loudly. Thirty-three frustrated professors is a pattern of complaints, not proof of decline. If the story stopped there, you’d be right to hold it at arm’s length.
And it goes further.
When quantitative data arrived, it was worse. The Nation’s Report Card—NAEP, the closest thing America has to a measurement of what students can actually do—released its 12th-grade results in September 2025. Reading scores fell to their lowest level since the test’s current form began: only 35 percent of high school seniors were proficient in reading, and 45 percent were below even the basic level. These are the students arriving on campus this fall. New results released this month show that younger students are recovering, while teenagers continue to slide. The anecdotes and the measurements now say the same thing: the students walking into freshman seminars are, on average, weaker readers than any cohort in decades—and the strongest readers are pulling further away from everyone else.
The workaround makes it worse. Faced with students who can’t finish books, colleges are choosing, mostly silently, to stop assigning them. Understandable. Also exactly backward. Deep and critical reading isn’t a prerequisite for college; it’s the product of it. The cognitive scientist Maryanne Wolf has spent a career showing that the reading brain is built, not born: the circuitry for sustained, analytical, empathetic reading forms only through practice and dissolves without it. Shrink the syllabus, and you’re not accommodating the deficit; you’re institutionalizing it.
And the timing could not be more perverse. This is happening at the precise moment AI makes shallow reading worthless. The one capacity that can’t be automated—sitting with a difficult text, tracking its claims, testing its evidence, evaluating its reasoning—is the one colleges are quietly removing from the curriculum because it became hard to teach.
There’s a counter-signal worth watching. Princeton just assigned Wolf’s Reader, Come Home—a book about the loss of deep reading—to its entire incoming class. Read that move for what it is: one of the most selective universities in the world telling its newest students, before they unpack, that the skill most at risk is the one it considers foundational. Princeton can see the same NAEP tables as everyone else. The difference is the response: name the problem and teach into it, rather than shrink the syllabus around it.
Here’s my claim, stated plainly. The students are not broken. They are unpracticed, in exactly the way a generation raised on skimming, scrolling, and now summarizing by machine would predictably be. A skill nobody teaches doesn’t get learned; that’s not rocket science; it’s a syllabus decision. Which means the crisis is also an instruction problem, and instruction problems always have answers.
That’s what the method is for. CERIC (Claim, Evidence, Reasoning, Implications, Context) is a structure for the thing students were never taught to do: stay inside a text long enough to judge it. It works on a chatbot’s answer (we will do this later this week), it works on a news story, and it works on the books that have quietly vanished from the syllabus. In coming posts, I’ll show how it works in an actual classroom, including with students who arrive saying they’ve never finished a book.
Colleges can keep shrinking the reading until there’s nothing left to assign. Or they can decide that teaching people to read deeply is what they meant to do all along.
Next week — CERIC vs. the Machine: The five-question method, run head-to-head against the chatbot that’s replacing the reading.
Read deeply,
Dr. Genevive
If you teach, I’d like to hear it straight: What’s happened to the reading load in your courses over the past five years? Please reply or comment. Anonymized patterns may show up in a future post.
Sources referenced: