- Scott H Young - https://www.scotthyoung.com/blog -

Why You Should Read Textbooks

Okay, file this piece of advice in the pile that nobody is going to follow (even though they probably should): you should read more textbooks.

Let’s assume for a second that you’re one of the few people who does read to learn more about the world. Let’s also assume that you’re interested in topics that are heavily researched: finance, health, nutrition, science or psychology. This probably eliminates most people, but I’m guessing as a reader of this blog you’re more inclined to such intellectual pursuits.

Ask yourself where you get information about these topics. Blogs? News? Popular non-fiction books?

There’s nothing inherently wrong with these sources. Some blogs and popular non-fiction books are crap—but many are not. Sometimes sacrificing empirical rigor can also be useful if the content is more pragmatic or impactful.

But if you do care about a subject, it probably makes sense to read at least one general textbook on it. That textbook may not fill you with the detailed knowledge of a PhD, but it can give the foundation for evaluating many other ideas.

Why Not Textbooks?

The value of reading a textbook (or, better, doing an online course [1]) is that it gives you a baseline for examining other aspects of that field. Taking one physics course would be enough to know why perpetual motion machines are scams.

Similarly, if you’re going to read books on the financial crisis, political blogs or start investing money—maybe it makes sense to have read one book on basic economics. I find it baffling that people have complex economic and political philosophies but haven’t learned concepts like supply and demand.

Ditto for psychology. One psychology textbook will hardly make you an expert. But it will at least make you aware that truths can’t be concluded from a single study, or that generalizing from a very narrowly designed experiment is dangerous.

The point of reading at least one textbook is to give an awareness of (a) the fundamental concepts most people agree with in a field and (b) where experts disagree.

Opinions and Experts

This blog is my opinion. That’s not necessarily a bad thing, as long as you realize that’s what it is. I’m not infallible, so there are probably quite a few opinions I’ve shared that are false.

I read other blogs that are mostly opinions. I like those blogs because they challenge me to think about topics, or introduce me to ideas I wouldn’t previously considered. As such, I try to strive to do the same in my own writing, open up new questions rather than just provide answers.

This is also true for the things where I’m an “expert”. I write about learning methods based on my experience and from working with students. I also try to use the science as best I can to guide the methods that I then test extensively.

Even then, I’m probably wrong about at least a few things. This is why I strive hard to push my own knowledge on the topic so that I am constantly adjusting or reevaluating past ideas.

The Danger of Only Using Secondhand Expertise

As a blogger, however, I’m also guided by other constraints. I need to write things people want to read. I would never write anything I knew to be intentionally false or misleading, but sometimes that means I write less about a topic that is boring, even if it is equally important.

For example, I consider doing practice questions with solutions to be one of the most important methods for learning technical subjects. I’ve stated this [2]before, but there really isn’t much more to that. Just do a lot of practice questions.

My methods like the Feynman technique [3], metaphors, visualization are things I spend significantly more time covering because they are unusual to students. Despite that, they probably only took up about 20% of the time during the MIT Challenge [4], in comparison to about 40-50% doing practice problems.

All writers face these constraints. Science journalism tends to hype results more than the research would warrant. Pop-psych books tend to make the field appear more unanimous in opinion than it really is. Bloggers categorize the unusual or interesting details first.

Reading a textbook, which is less influenced by these constraints, can give you some awareness of these biases and correct for it in your thinking. I won’t stop reading blogs or popular books—textbooks are dry and often impractical—but having knowledge of one or two helps me balance some of the biases inherent in popular writing.

Degrees of Belief

I’ve written before that the only appropriate way to look at knowledge is through degrees of belief [5]. This means that almost nothing (aside from logical truths) is known perfectly. Instead everything is known to different degrees of certainty.

Scientific principles like relativity are so well-established we can safely say they are correct to some minuscule measurement with enormous accuracy. That’s enough to warrant rounding down that doubt to zero for most situations.

Higher-order theories in social sciences or popular opinion have more doubt built in. That doesn’t mean they need to be rejected, simply that you give yourself more room to reject them later if better theories are generated.

For now, I’m confident in the learning techniques I use, but I’m always looking for better models that might have more evidence and therefore better reliability. The hard part is realizing that this is an ongoing process. You can never just put your hands up and say, “Done!”

Different sources of information have different degrees of evidence as well. A blog article providing an opinion has significantly less evidence than dozens of controlled, well-repeated studies on a particular fact. When the two directly conflict, I side with the research.

However, often the research isn’t in yet. In these cases, I enjoy others’ opinions since it lets me entertain speculative theories while also allowing me room to continue investigating. I’ve enjoyed all of Malcolm Gladwell’s [6] books, but it would be ridiculous to assume that he isn’t making any assumptions or leaps to stitch together a cogent narrative. I’m willing to accept this uncertainty, but I wouldn’t mistake it for fact.

Thinking in degrees of belief is not an easy task. I also understand the attitude that we need to draw a line somewhere, above which all facts are compelled to be believed, below which anything can be safely ignored. But, ultimately I think this is a weak position as well. It is often abused to allow you to accept some opinions but not others with exactly the same volume of evidence.

Well-Rounded Knowledge

My advice is to read one textbook on a subject for every 4-5 popular books or 50-100 articles you read about it.

Reading only textbooks is probably impractical. I want knowledge not just for knowledge’s sake, but to do something useful with it. Reading a book about exercise that distills research into practical tips is probably more useful than a textbook in physiology. Same for personal finances, learning, productivity or nutrition.

Reading only popular nonfiction is probably misleading. If you’re going to read a dozen books in personal investing, it probably makes sense to at least understand the rationale behind the efficient market hypothesis. Some authors will do this for you, but a lot of it won’t because of the constraints mentioned earlier.

Picking Textbooks

I’ve used the word “textbook” here loosely, but broadly I’d say it means two things:

  1. The book tries to describe established viewpoints, rather than argue for a particular one (except where there is consensus). Academic textbooks are good for this because universities usually try to pick books without any severe bias.
  2. The book focuses on fundamental concepts necessary to understand the field, not just minor details or conclusions. A good textbook should teach you how to think about a field, not just what to think.

Since textbooks are rarely the trending topics on social media, finding good ones comes down to searching for them. Look for ones that have good Amazon reviews from researchers in the field, or ones that are used in classes at major universities. Older editions are often better, because you can get them used for cheap.

Have you read any textbooks that you felt were engaging and informative about a topic you’ve studied? Please share them in the comments [7]!

Edit: March 6, 2013 – Reader, Luke, has posted a link to a fantastic resource for finding great textbooks. Check out LessWrong’s best textbooks on every subject [8].