ybooks

The case

Why read books

There are two kinds of answer to this question and they should not be mixed. One is an argument about what a book is — no study required, and none offered. The other is empirical: things researchers have measured about people who read. Both are below, kept apart on purpose.

The argument

Nothing in this section is a scientific claim. It is the case as anyone would make it over a long dinner, and it is the half that actually persuades people.

A book is the highest-bandwidth channel into another mind

Someone spends five years — sometimes forty — thinking about one thing, then compresses it into fifteen hours you can carry in one hand. Nothing else in culture has that exchange rate. A conversation with the same person would take a decade to cover the same ground, and they would not have had time to get it right.

You get the reasoning, not just the conclusion

A feed hands you the takeaway with the argument stripped out, which is exactly the part you need in order to disagree with it intelligently. A book has room for the objection, the counterexample, and the place where the author is unsure. That is not padding. That is the thing you came for.

Some ideas only fit at book length

There are thoughts that cannot be posted. They need two hundred pages of groundwork before the point lands, and an idea that survives its author having to defend it for two hundred pages is a different class of object from one that fit in a paragraph.

Books are the cheapest way to talk to the dead

Almost everyone worth learning from is already gone. Reading is the only technology that gets you into the room with them anyway — and it works in both directions, which is the whole reason we know anything at all.

Sustained attention is a skill, and it is use-it-or-lose-it

Following one argument for six hours is a capability, not a personality trait. People who stop practising it report losing it, and people who start practising it report getting it back. A book is the standard piece of equipment for that particular exercise.

You choose the book. The feed chooses you.

Every other reading surface you touch was assembled by a system optimising for something that is not your interests. A book is the last widely-available medium where you pick the next thing, on purpose, and nobody gets paid when you keep going.

The evidence

Every finding here is real, published and linked. Every one also comes with the reason it proves less than a headline would tell you. Reading is extraordinarily hard to study: you cannot randomly assign people to a life of books, so nearly all of this is observational, and people who read a lot differ from people who don’t in a hundred ways researchers cannot fully adjust away.

20%lower mortality risk over 12 years

Book readers outlive non-readers

Yale researchers followed 3,635 adults over 50 for up to 12 years. Book readers had a 20% lower risk of dying during the study than non-readers (hazard ratio 0.80). Even under 3.5 hours a week showed the effect, and books beat newspapers and magazines — reading a book was doing something reading a periodical was not.

The honest caveat. Observational. Adjusted for age, sex, race, education, wealth, comorbidities, self-rated health, marital status, depression, vision and employment — and after that adjustment the raw 23-month survival gap shrinks to about 4 months. The risk ratio is the robust finding; "reading adds two years to your life" is not what this says.

Bavishi, Slade & Levy, "A chapter a day: Association of book reading with longevity", Social Science & Medicine 164 (2016)

32%slower late-life cognitive decline

A reading life ages better than the brain it sits in

Rush University tested 294 people annually until they died, then examined their brains. Frequent readers and writers declined about 32% more slowly than average; the least mentally active declined about 48% faster — and this held after accounting for the physical damage found at autopsy. Two people with the same plaques and tangles did not end up in the same cognitive place.

The honest caveat. Reading was measured together with writing and other cognitive activity, so this is not a books-specific result. Reverse causation is live: an early decline can reduce reading rather than the other way round.

Wilson et al., "Life-span cognitive activity, neuropathologic burden, and cognitive aging", Neurology 81(4) (2013)

80books at home is where most of the gain lands

Growing up around books shows up in adults decades later

Across 31 societies, the number of books in the home at age 16 predicted adult literacy, numeracy and digital problem-solving — over and above the parents’ education and the person’s own schooling and job. A teenager in a bookish home who left school early ended up roughly as literate as a university graduate from a bookless one. The curve is steepest early: going from almost none to about 80 books buys most of the effect, with gains continuing to around 350.

The honest caveat. Home library size is also a proxy for things it is hard to measure — parental attitudes, family stability, income beyond what the controls capture. Buying 80 books is not the intervention this study tested.

Sikora, Evans & Kelley, "Scholarly culture: How books in adolescence enhance adult literacy, numeracy and technology skills in 31 societies", Social Science Research 77 (2019)

g ≈ 0.15small but real effect on social cognition

Fiction moves the needle on reading other people — a little

A meta-analysis of 14 studies and 53 effect sizes found that reading fiction produces a small, statistically reliable improvement in social-cognitive performance compared with reading non-fiction or nothing, and the effect survived tests for publication bias.

The honest caveat. Included deliberately as the honest version of a claim you have seen oversold. The famous 2013 Science study — read a few pages of literary fiction, score better on a theory-of-mind test — did not replicate cleanly, and later preregistered attempts were mixed. "Small and real" is the defensible claim; "fiction makes you empathetic" is not.

Dodell-Feder & Tamir, "Fiction reading has a small positive impact on social cognition: A meta-analysis", Journal of Experimental Psychology: General 147(11) (2018)

g = −0.21paper beats screen on comprehension

The format is not neutral

Fifty-four studies, more than 170,000 readers: comprehension was measurably better on paper than on screen. The gap grew across the years studied rather than shrinking, and "digital natives" did not do better than people who grew up on paper — so this is not something the next generation of hardware fixes.

The honest caveat. The effect is for expository text, and is largest when reading is under time pressure. For narrative — novels — a separate meta-analysis found no screen penalty. Read fiction on whatever you like; sit down with paper for the hard stuff.

Delgado, Vargas, Ackerman & Salmerón, "Don’t throw away your printed books: A meta-analysis on the effects of reading media on reading comprehension", Educational Research Review 25 (2018)

What this site refuses to claim

A page that prints only the flattering numbers is an advertisement. These three claims are everywhere, they would help the argument, and they are not good enough to use.

"Six minutes of reading reduces stress by 68%"

Traced to a 2009 University of Sussex press mention of work by Dr David Lewis. No peer-reviewed paper was ever published, no method or sample was released, and the number has been copied between articles ever since without anyone reaching the study. It may well be true. Nobody can currently check.

"Reading literary fiction immediately makes you better at reading emotions"

The 2013 Science experiment behind this failed to replicate in an independent attempt, and the authors’ own preregistered follow-ups came out mixed. The surviving version of the claim — a small average effect from fiction reading — is on this page, with its actual effect size.

"CEOs read 50+ books a year and that is why they are CEOs"

Self-reported, unverifiable, and backwards even if true: people with staff and control over their calendar can read more, which is a consequence of the job rather than a cause. What successful people actually read, in their own words, is on the reading page — as quotes with links, not as a productivity claim.

The bottom line

The measurable effects of reading are real and modest. The unmeasurable one is larger and is the actual reason to do it: a book is the only way to spend fifteen hours inside a mind that is not yours, at your own pace, with the right to stop and argue. Everything else on this page is a footnote to that.

What notable people say about it → What they are reading now →

That little bar

It says how long ago something was said, so you can tell last month from 2014 without doing the arithmetic. Full and green is recent; it drains and cools as the date gets older, and empties completely at about 6 years.

A striped bar means the date itself is an estimate — the source did not carry an exact one. An entry with no date at all gets no bar.