People often quote the line, “Live and learn.” Most take it as a statement about lifelong learning, and I agree with that—but I also think it points to a way of living, something closer to constant self-examination. The two overlap, but they are not quite the same.
Learning, when it concerns knowledge or skill, usually moves from study to practice: from imitation to use. Reflection on life works in the opposite direction: it starts with experience and extracts knowledge from it. Whether the old saying originally meant one of these or both does not matter much to me. When I talk about learning here, I mean both together.
Knowledge is not the same as technique
When people talk about learning, they often assume the object of learning is simply “knowledge.” I prefer to separate two things: knowledge and technique.
Knowledge is a system of understanding built through experience, research, inference, and reflection. It is theoretical, abstract, and structured. It answers questions like what is this and why is it this way. It usually appears in the form of concepts, principles, and theories.
Technique, by contrast, is the practical application of knowledge. It is about methods, tools, procedures, and systems designed to achieve a specific goal. It answers how to do it. It is operational, repeatable, and tied to concrete use.
The difference matters:
- Knowledge emphasizes underlying logic and explanation.
- Technique emphasizes execution and results.
- Knowledge tends to travel across fields more easily.
- Technique is often highly dependent on context.
- Knowledge changes relatively slowly.
- Technique can update very quickly.
A theory in philosophy or psychology belongs to knowledge. Writing Python code, making SQL queries, or building a chart in Tableau belongs to technique.
For the rest of this piece, when I say “learning,” I am mainly talking about learning knowledge rather than technique.
The three main paths to knowledge
As I see it, there are three broad ways to learn knowledge:
- books
- papers
- the information scattered through ordinary life
Books
Of the three, books usually provide the fullest explanation, the most continuous logic, and the most internally coherent account. That is why I still regard books as the best path to knowledge.
But “best” does not mean flawless.
Books have several weaknesses:
- They may be old enough that some of their claims are outdated.
- Because they aim at completeness, they often contain too much at once, which raises the difficulty of learning.
- Authors have their own commitments and biases, so a book does not present absolute truth so much as a truth shaped by its author.
- Different books can contradict one another while each remains internally consistent on its own.
- Many books are deeply marked by the era in which they were written, which gives them both historical value and historical limits.
That last point is especially important. A book may make perfect sense inside its own time and still confuse or mislead a modern reader who lacks the historical setting needed to understand it.
Take Plato’s Republic. Without some understanding of the Greek city-state world and the material conditions of that age, it is hard to understand why his ideal state looks the way it does. Read in isolation, it can seem entirely at odds with modern ideas such as equality, freedom, or sexual fidelity. Later readers who copied such visions too mechanically produced results that were deeply damaging.
Another example is the contrast between Karl Marx and Adam Smith. They lived in the same broad European age, and both closely observed capitalist production, especially as it developed in Britain. Yet Marx concluded that capitalism was exploitation, while Smith concluded that it could generate the greatest public benefit through self-interest and markets.
Which one was right? That question has troubled countless readers, and there is still no simple final answer. Even experts hesitate. When I reread a respected complete translation of The Wealth of Nations, I noticed that the introduction cited Marx’s criticism of Smith. That may seem unnecessary if one supports Smith’s free-market position, but to me it reveals how strongly intellectual life is shaped by its time. Looking back now, how many people can say with certainty that free-market theory is wholly right and socialism wholly wrong, or the reverse? History often seems to answer differently: each contains something valid, and reality forces combinations rather than pure choices.
That is exactly where beginners get into trouble. If you believe Smith too early, you may dismiss Marx too quickly. If you believe Marx first, you may reject Smith just as blindly. Without a broad enough understanding, people choose according to temperament, and that tendency can easily become an intellectual cocoon. Ironically, the more authoritative a book is, the more complete its argument, and the tighter its internal logic, the easier it is to submit to it too fully. That is the danger of shallow reading and of believing books without resistance.
Even with all those problems, I still think books remain the strongest foundation for learning.
Papers
Papers are much shorter than books, so they usually focus on only one issue, or even one narrow point. Their strengths are precision, sharpness, and timeliness.
That also limits them.
Papers are excellent for learning a specific idea, but poor for building a whole framework from scratch. Different papers often address scattered questions. If the reader does not already have a structure in mind, reading more papers can actually produce more confusion, because papers also contradict one another all the time.
I suspect this is one reason many people learn from fragmented online commentary and end up more confused than before. It is hard to judge whether a paper’s argument is right or wrong if you only look at the paper itself. A well-argued article is not easy to refute unless you already possess some larger framework.
For example, I recently saw a well-known economist argue against directly subsidizing citizens or distributing consumer vouchers during an economic downturn, suggesting instead that the better path was to increase investment, create jobs through investment, and let employment drive consumption. I disagree with that view. Both data and experience suggest that in China the effect of investment on GDP growth has already become quite weak. At the same time, the economy faces severe overcapacity and an urgent need for industrial upgrading, neither of which can be solved simply by investing more. And under the pressure of artificial intelligence, investment in high-tech sectors does not necessarily solve the employment problem. If jobs are not solved, then “investment driving employment” becomes much less convincing.
Papers may be fragmented, but they have one major advantage: they are current. In that sense, they sit somewhere between knowledge and technique. Once a person already has some intellectual framework and wants to test, update, or apply it, papers become extremely useful.
Information from everyday life
The third path is easy to underestimate: the information carried by ordinary life.
Novels, films, news reports, games, casual conversations, travel, workplace observation—all of these can contain knowledge. The problem is not absence but disorder. Life produces an enormous amount of information, but it is scattered, mixed with noise, and rarely presented in a clean conceptual form.
A novel about business rivalry can be mined for management ideas. A story about the bureaucracy of an earlier dynasty may reveal something about historical selection and promotion mechanisms in official life. News, television dramas, and even games can become sources of knowledge if one knows how to extract patterns from them.
In that sense, learning from life is a kind of self-discovery. If you can pull systematic understanding out of chaotic experience, you are not just receiving knowledge—you are producing it. People who have seen more, lived more, and endured more are often better at this. That is one reason people say that age makes a person shrewd.
The principle is simple enough. Books themselves often rely on case studies, and those cases come from life. Many conclusions generated through big-data analysis are also abstractions drawn from massive amounts of everyday information. The logic is the same as a person learning from accumulated experience, only at a different scale.
So yes, knowledge is available everywhere, as long as the mind keeps working.
But this path is inefficient. A book may teach a principle in a few sentences; discovering the same principle on your own may take half a lifetime. That helps explain why the old idea of “traveling ten thousand miles is like reading ten thousand books” sounds admirable but is less often practiced now. In an age overflowing with organized knowledge, extracting it directly from life is no longer the fastest method.
That does not mean learning from life is unimportant. Quite the opposite: I think it is necessary.
Knowledge is an abstraction from experience, much as sea salt is refined from seawater. The body needs salt, but we do not drink seawater to get it; we consume a small amount of salt instead. But if we need vitamins, we do not live on vitamin pills alone and forget oranges or tangerines.
Learning knowledge is meant to enrich life, not replace life. The point of being alive, and of pursuing oneself seriously, is to deepen one’s experiences and perceptions. Knowledge is an efficiency tool. It should not be mistaken for the thing it is meant to serve.
That is why I also oppose the kind of person who does nothing but grind through books and never learns how to live.
How I rank these paths now
If I had to place them in order for myself, I would rank them like this:
books, then papers, then life information.
That ranking is personal, not universal. I choose it because I have already crossed the threshold of reading difficulty in the areas I care about, so higher-efficiency methods suit me better now.
But each person should make their own choice. Top-down and bottom-up approaches can both work. Someone might start with science fiction, then move into popular books on quantum physics, and only later read about parallel-universe theory more systematically. That is a perfectly valid route.
For books, I would suggest a progression like this for beginners entering a field:
- start with general introductions
- move to more narrative-style professional books
- then read recent specialist works
- only after that, read the classics broadly
This makes entry easier, progress smoother, and eventually helps build both a knowledge system and a judgment system.
As for reading format, I now mix print and digital. I buy physical books only when they are dense, difficult in language, or classic enough that I want to keep them. Everything else that can be read quickly goes on an e-reader.
Digital access is also convenient and cheap. A reading membership through WeChat Reading can be obtained through check-in activity, with 13 months costing only 50 yuan. JD Reading is bundled with a JD membership. Between those two, I can usually find more than 80 percent of the books I want.
For papers, I have spent a long time looking for a good professional aggregation platform and have found that harder. Caixin is very good, but the subscription is beyond what I want to pay. Other outlets each have one problem or another. So far, I have found two useful and free ways to access papers and reports. One is the Choice client from East Money, which includes news, papers, professional reports, and database queries; it is broad and practical. The other is an account called “Principles of Economics” on Folo, which updates roughly two economics papers per day and generally selects material well enough that my time is not wasted.
Given the amount of information online today, whatever your intellectual interest is, there is probably a good aggregation point somewhere. Whether you happen to find it is partly a matter of luck.
As for learning from life, everyone has their own method, but I would still advise aiming for efficiency. More importantly, you need to know what you are trying to understand rather than grabbing whatever happens to pass by.
If you want to know how people feel about China’s basic medical insurance system for rural and non-working urban residents, then the useful thing is to talk to many people about that topic and guide them toward questions you have deliberately prepared. It makes little sense to spend two years learning cooking from a chef if your real question lies elsewhere.
One experience left a deep impression on me: I traveled to Tibet once, and that single trip gave me a deeper understanding of Tibetan Buddhism than all the films and books I had encountered over much of my life—not only about Tibetan Buddhism, but in some ways about Buddhism more broadly. The same is true of landscapes. Walking through a place once can teach more than hundreds of documentaries.
From that I drew a practical rule: when a subject is heavily colored by bias or strong interpretation, direct experience is often better than learning through second- or third-hand accounts. On the other hand, if a subject is less entangled with personal emotion, or if direct observation is unrealistic, then books, papers, and other mediated forms are the better choice.
My own path into learning was slow and accidental
Many people say they want a life of learning, but the actual process differs wildly from person to person. Those different paths create different experiences, and eventually different long-term habits.
My interests are fairly narrow at the moment: economics, sociology, philosophy, history, and general science reading. But I no longer resist learning, and I can now read books and papers with a calm mind rather than defaulting to short videos and noisy self-media platforms.
A major reason for that change was the pandemic period.
A few years before COVID, I spent about a year and a half playing the game Onmyoji. In the end I gave it up after my wife kept nagging me about it, and I started using most of my free time for reading instead. After the pandemic began, staying at home gave me even more free time. I was not very interested in film, television, or self-media, so I read a large number of books and gradually pushed through the hardest beginner stage.
The beginning of learning is painful in much the same way as the beginning of many worthwhile habits. At first there is resistance, even a kind of rejection. You need interest, willpower, or some outside force to carry you through that stage. Only after that can learning become self-reinforcing.
When I first started reading economics, I took the wrong route. With no one to guide me, I began directly with the classics. The result was predictable: I understood little, felt lost, and became intimidated by many books because so much of the content made no sense to me.
Fortunately, I then read two histories of economic thought. Once I had a sense of how economic ideas developed over time, the classics became far more readable. I also happened, by good luck, to buy works by Taleb and Shiller. Their accessible style made me genuinely interested in economics books. The luckiest break of all was reading Steven Cheung’s Economic Explanation, which gave me a solid framework for my economic thinking. From there, doors opened one after another: Friedman, Keynes, Hayek, Schumpeter, Mises, Menger, Smith, Fisher, Ricardo, Hume.
Then economics led naturally into other fields. To understand many economic questions, I needed historical context, so I read more history. European history led me into religion. Religion led into Christian theology. Theology made philosophy unavoidable. Philosophy pushed me to learn some logic. To better understand economics, history, and philosophy, I also picked up some sociology.
There was a visible thread connecting all of this, so once the interest was there, learning no longer felt painful. After gaining some basics and building a web-like structure across subjects, I became able to choose suitable books on my own and continue reading naturally.
Before that habit formed, however, I had gone through a long period of chaotic, trial-and-error reading. After frustration, I would retreat either to novels and lighter material or to endless internet browsing and online games. That taught me something simple: for people who are not born with strong curiosity or self-discipline, lifelong learning usually needs a trigger.
What I really mean by “learning” here
I wrote these reflections casually rather than as a grand planned essay. They came partly from thinking about education and partly from watching how friends and relatives approach learning and life planning.
I separated knowledge from technique because, for most people, learning is driven by practical pressure. It is utilitarian. Technique often improves life quickly and visibly. I have never focused much on technical training, nor have I mastered any particular technique well. I am also not especially persuaded by a view of life centered primarily on making a living. So what I have written here is about learning knowledge, not about career skill-building.
I am not writing this because my learning outcomes are especially impressive. I am writing it because I am satisfied with my current state of learning. I suspect that kind of satisfaction—steady, calm, and self-sustaining—is what many people actually want.