The Ten-Book Rule for Smarter Thinking

Ten well-chosen books are usually enough to understand the expert consensus on any reasonable question you might have.

At this point, I can imagine the doctors, lawyers and anyone with a PhD cringing at my exceedingly low bar for expertise. But let me unpack that above rule:

• The books have to be well-chosen. Any ten random books on a subject won’t converge to the expert wisdom. Even a hundred books won’t if they’re low-quality. I’ll define high-quality books below, but this is an important caveat to remember.
• Understanding an expert consensus doesn’t make you an expert. Understanding knowledge is a much lower bar than creating knowledge. It’s also a lower bar than successfully applying knowledge to diverse domains. My argument isn’t that ten books would be enough to make you a cardiac surgeon, but they would be enough to understand what most experts think is the right way to do a coronary bypass.
• There must actually be an expert consensus on the question (or at least, a few dominant viewpoints). You can’t get an expert consensus if experts don’t actually agree on the answer. Similarly, if the question hasn’t been addressed because the fields choose not to represent questions that way, you might be out of luck.
• A reasonable question is down-to-earth. The highest levels of a field can often formulate questions a novice wouldn’t even think to ask. Understanding string theory or Continental philosophy often requires a much more extensive background of knowledge to even ask reasonable questions. But “why is the sky blue?” or “what is existentialism?” are definitely answerable within ten books. A question is reasonable when it is both something a layperson could readily formulate AND a good answer already exists.

Ten books is a substantial threshold in terms of casual interest. It’s much more than perusing a Wikipedia article or an essay. The books in question are not fun, easy-to-digest pop-science. Even at the reasonable pace of a book-per-week, this is about three months of work.

Ten books are considerably less than what it takes to become an expert in anything. But if you want to answer a reasonable question (that meets the criteria above), you can probably get a satisfactory answer just by doing the work.

Given the relatively low bar I claim is needed to understand an expert consensus, why don’t more people do this?

How to Pick the Right Books

The first difficulty people have with this approach to research is that they pick the wrong books.

There are three kinds of books that tend to slow the path to understanding expert consensus, and unfortunately, they’re also the kind that tends to line bookstores and best-seller lists:

1. Books with “new” ideas. Most ideas are old, even in supposedly cutting-edge fields. If a book is full of novelties, that’s another way of saying it is full of things yet to be widely proven.
2. Books with “useful” ideas. Pragmatism is a virtue, but it often distorts research results. Don’t confuse “what’s the right way to think about this issue?” with “what are practical things I can do about it?”
3. Books with “revolutionary” ideas. Heterodox books that explicitly frame themselves as a paradigm change will make it harder to understand the orthodox perspective.

This doesn’t mean the above books aren’t worth reading, just that they don’t count for the “ten” you need to understand an expert consensus. Reading ten self-help books, or ten books about the “new science of X,” or even a controversial best-seller that shows why all the experts are wrong may be fun and interesting, but it will only slowly get you to the general picture experts have about a topic.

What books should you read instead?

I would suggest three types of books, in the following order:

1. Up-to-date textbooks. Textbooks are one of the most valuable books to read because they are written to represent expert consensus. Even authors with strong heterodox opinions usually present a balanced picture in their authored textbooks.
2. Academic monographs. Monographs tend to be more focused than textbooks, so while you may not get a general survey of the field, you’ll often get closer to the answer you seek via a monograph. If good monographs don’t exist for the question you have in mind, then review articles are often a good substitute.
3. Canonical texts that the field cites as authoritative. I don’t usually start here because, as a novice, identifying these texts and understanding their significance is often tricky without greater context. However, when a particular work is oft-cited in textbooks or monographs, I try to fill in my understanding of it.

My claim with the above rule is that if you picked a well-posed question like, “how should I invest in the stock market?”, “what’s the best way to treat anxiety?” or “how do batteries work?” you’d get a good read on the expert consensus by reading those books.

Why Care About the Expert Consensus?

Why should we care about the consensus view anyway? Shouldn’t we care about the truth, even if that means turning away from the opinions of a bunch of ivory-tower academics? I think there are good reasons why understanding the modal opinion of experts is still very useful, even if it falls short of knowing the “truth”:

• In healthy intellectual fields, “expertise in X” is pretty close to “people who know a lot about X.” Learning the expert consensus is, therefore, a reasonable estimate of the answer to: “if I learned as much as an expert, what opinions would I likely form?” The ten-book rule helps you get close to this estimate.
• Discourse tends to be grounded in a consensus viewpoint. Therefore, it’s impossible to properly understand a heterodox view without knowing what it seeks to reject. Thus even if you strongly suspect that experts of a particular stripe have the wrong mental model, you still need to learn the consensus ideas and language to understand the alternatives.
• Knowing the “truth” is problematic; knowing the expert consensus is achievable. Without wading too far into epistemology, there are well-known difficulties in acquiring reliable knowledge about the world. Practically speaking, every discipline has its own standard of evidence and methodological techniques. In contrast, figuring out what experts tend to think is eminently achievable and doesn’t fall into the same quandaries.

Engaging in More Research Projects

In keeping with my previous post, I think there are two broad ways to learn more about the world: building up from the basics, or learning for specific ends. Both have merit, but after you have mastered the basics, the sheer volume of knowledge explodes, so it helps to ask more pointed questions.

Self-conducted research isn’t without pitfalls. As mentioned above, a major reason people don’t reach the expert consensus after ten books isn’t that their goal was impossible. It’s because they picked the wrong books. Similarly, online sleuthing often leads one further away from reality as bogus sources and “alternative” accounts drown out any reasonable interpretation.

However, I tend to think that these problems have less to do with critical thinking and more to do with motivated reasoning. If you genuinely want to know what experts think about a topic and are willing to read at least ten serious books about it, I would wager you’d be on-target more often than not. All that’s required is to put in the effort and actually want to know the answer.