Source: Thought Steel Imprint
1. From Hume to Buffett
Of the following three options, which one do you prefer?
Option 1: For a stock you are very bullish on, would you—
A. Wait for a lower price, at the risk of missing out on a strong performer
B. Buy as long as the price is reasonable—strong performers are rare and should not be missed
Option 2: Which segment of the technology sector do you prefer?
A. AI computing power, which has real market demand, a clear technological roadmap, and reliable orders
B. AI applications, which currently lack viable products and genuine market traction, but are backed by belief in the broader industry trend
Option 3: Someone introduces you to a stock-picking system that has selected several stocks with strong gains over the past six months. However, this time it has picked a stock you don’t agree with. What would you do?
A. Trust your own experience rather than the stock selected by the system
B. Trust the stocks selected by this system rather than your own experience.
If you chose mostly option A, your thinking style leans more toward 'empiricism' from the epistemological perspective of philosophy; if you chose mostly B, it leans more toward 'rationalism.'
When investing, one should ask several 'whys,' but at the end of a chain of 'whys' lie philosophical questions.
For example, Buffett and Munger share similar investment philosophies, yet they still disagree on many specific issues. Some stocks Munger selects, such as Alibaba, are not fully endorsed by Buffett. As the saying goes, 'Gentlemen harmonize without conforming.' Laypeople focus more on 'harmony,' whereas investors should pay closer attention to 'differences'—as these differences offer an opportunity to re-examine one’s own investment framework.
Buying individual stocks means buying a company's future; buying an index means buying into economic expectations. Since the future has not yet occurred, it inherently carries uncertainty. Understanding this 'uncertainty' is fundamentally an issue in epistemology—a philosophical field historically divided into two major schools of thought. To grasp the origins and trajectory of your own research inclinations, you must first understand the distinctions between these two schools.
2. Rationalism: Explaining everything
If someone tells you they have discovered a magical set of indicators capable of predicting stock price movements with high accuracy, there are two possibilities: either they are a 'fraud,' or they are a genuine 'rationalist' who adheres to rationalism in epistemology.
The pioneer of 'rationalism' was Descartes, who famously—but often misinterpreted—stated, 'I think, therefore I am.'
Many interpret this statement as meaning 'the value of one’s existence lies in thinking,' which sounds inspirational. However, Descartes’ actual intent was that 'thinking is the evidence of my existence.'
You might wonder: why would my existence need evidence? After all, I can breathe, see, hear, and speak—how could that be false? Yet Descartes argued that even 'my existence' could be an illusion (as illustrated in The Matrix, the 'brain in a vat' thought experiment, or artificial intelligence scenarios). Everything must be verified through rational methods.
This is the core tenet of 'rationalism': knowledge primarily originates from rational deduction and innate ideas rather than sensory experience. Consequently, rationalists also tend to favor 'metaphysics.'
Metaphysics is a methodology contrasted with 'physica' (or 'the study of the concrete'). For instance, in understanding '1+1=2,' the 'physica' approach involves using numerous concrete examples—such as 'one cup plus another cup equals two cups'—to grasp the concept. In contrast, metaphysics considers such exemplification unscientific, as it cannot exhaust all possible instances. Instead, it insists on transcending specific forms and proving '1+1=2' through mathematics alone—hence the term 'metaphysics.'
Another major figure in rationalist philosophy was Spinoza, whose seminal work, Ethics Demonstrated in Geometrical Order, addressed ethical questions concerning God, the mind, emotions, and more entirely in the form of mathematical proofs.
Rationalist epistemology reached its zenith following Newton’s laws of motion and the Industrial Revolution. People came to believe that all phenomena could be governed by a few simple physical laws, much like mechanics. Both humans and the universe were viewed as machines: given the current state and a set of laws, the future could be precisely predicted. Rationalism thus evolved into mechanistic determinism.
In Isaac Asimov’s renowned science fiction series Foundation, there exists a discipline called 'psychohistory,' which treats the socioeconomic and political activities of populations across more than 20 million planets in the galaxy as historical big data. From this, universal laws are derived to forecast the future trajectory of human society—including the fall of the Galactic Empire and the path to its salvation.
However, the two World Wars of the 20th century brought humanity down from its hubristic peak. Classical physics was revealed to be only partially true, and human rationality itself came into question—perhaps innate rationality did not exist at all. Throughout the 20th century, both science and art were infused with strong elements of irrationality.
Nevertheless, rationalism remains the foundational methodology of modern education. In subjects like mathematics, physics, and chemistry, theories are first rigorously proven and then applied to explain the world. Even in the humanities—history, literature, and philosophy—a broad theoretical framework is established first, then used to interpret all related phenomena.
Investment theory is no exception; mostTechnical Analysisapproaches, whether Elliott Wave Theory or candlestick charting techniques, aim to construct an all-encompassing explanatory framework and are heavily imbued with rationalist elements.
Among various investment theories, Chan Theory exhibits the strongest rationalist orientation. Its analytical method is thoroughly Cartesian: it systematically categorizes all market structures and price movements. The notion that a system equals the sum of its parts reflects classic mechanistic thinking. Moreover, the core tenet of Chan Theory—'all trends ultimately achieve perfection'—is virtually a restatement of determinism.
Early value investing was also deeply infused with rationalism. In his 1933 book Security Analysis, Benjamin Graham pioneered the transformation of corporate financial statements into measures of investment value. Navigating the labyrinth of accounting figures—assets, liabilities, cash flows, and profits—he reconstructed a company’s intrinsic worth. Unlike later value-investing books that emphasize philosophy over methodology, Graham’s work presented a complete and systematic framework for valuation. Consequently, Graham is not only regarded as the father of value investing but also as the 'father of the CFA (Chartered Financial Analyst) designation.'
Methodologies in macroeconomic analysis and investment are even more saturated with 'rationalism.' Macroeconomics, as a social science, has always been the most mathematically intensive, and 'explaining everything' has become the starting point for complex financial models.Merrill Lynch Investment ClockIt established a comprehensive framework for major asset rotation based on interest rates, growth rates, and risk appetite. Although its explanatory power regarding real-world phenomena is limited, the aesthetic appeal of its classical theoretical constructs is truly admirable.
In summary, rationalism in investing exhibits at least two key characteristics:
1. It strives to construct a perfect system capable of explaining the patterns governing markets or stock price movements;
2. Its reasoning process places greater emphasis on 'metaphysical' logical deduction rather than merely settling for 'empirical' observation-based generalizations.
Therefore, fields such asTechnical Analysisdata-intensive macroeconomic analysis and investment are more prone to 'rationalism.' In value investing, only Benjamin Graham exhibited strong 'rationalist' tendencies—his assessments of business value were all derived through calculation.
However, research in investing differs from other forms of research: it must not only possess explanatory power regarding reality but also reliably predict future profitability—and generate repeatable returns. This is precisely the fatal flaw of most rationalist approaches.
Most value investing masters focus instead on examining the actual operational conditions of businesses—a hallmark of another major epistemological school: empiricism. Let us now return to the era of Descartes.
3. Empiricism: Experience teaches better than instruction.
Although 'rationalism' venerates logic and deduction, the foundational elements of its theories invariably rest on unprovable assumptions—either a few self-evident 'axioms' or presuppositions derived from 'innate reason.'
For instance, all technical analysis theories originate either in metaphysical concepts like yin-yang and the Five Elements or in psychology, which itself is derived from observed life experiences. This resembles a grand and sturdy castle built upon a foundation that is essentially an unexplained 'black box.' Consequently, even among adherents of 'rationalism,' entirely opposing conclusions can be logically deduced—a key reason why empiricists dismiss rationalism. Empiricists contend that truth can only arise from experience.
Francis Bacon, a leading philosopher of empiricism, famously declared, 'Knowledge is power.' Like his broader philosophy, this statement carries no ambiguity; here, 'knowledge' refers specifically to insights derived inductively from observable 'facts,' rather than those deduced purely from abstract 'theory.'
Thus, Bacon’s core methodology was inductive reasoning, emphasizing careful observation of real-world phenomena and subsequently elevating these observations into general theories.
Empiricist thinking aligns more closely with how most ordinary people reason. As the saying goes, 'Experience teaches better than instruction.' Telling a child a thousand times in theory that 'fire is dangerous and must not be touched' is far less effective than letting the child experience it firsthand.
However, inductive reasoning has a critical flaw: since humans cannot observe every possible instance, conclusions drawn through induction are not necessarily valid. For centuries, everyone believed all swans were white—until someone discovered the 'black swan.'
This weakness of inductive reasoning is nearly as pronounced in investing as its strengths. Lessons learned from one stock may become entirely irrelevant for another, making investors reluctant to rely confidently on inductive methods.
Many argue that empirical rules do not need to be 100% accurate—being mostly correct is sufficient. Yet in investing, 'high probability' can be especially perilous. The more strongly one believes in a pattern, the larger the position one tends to take; if that belief proves wrong, the resulting loss may be irrecoverable. Consider investors who still firmly believe in Vanke—you cannot definitively say they are mistaken, yet their conviction carries significant risk.
In summary, empiricism is highly practical in everyday life, but in more sophisticated domains such as investing, many seek more robust methodologies—precisely the niche from which the 'market' for 'rationalism' emerges.
In fact, philosophers have long offered a more insightful response to the shortcomings of inductive reasoning—namely, skepticism, which lies at the root of Buffett’s thinking.
4. Skepticism: All causality is illusory.
Starting with Buffett, most value investing masters are empiricists, yet their empiricism is deeply infused with skepticism—an intellectual lineage that traces back to the British philosopher David Hume.
Hume’s skepticism was later distilled by Bertrand Russell into the 'Farmer Principle,' a concept that gained widespread recognition after being referenced in the novel The Three-Body Problem.
Every time the turkey on the farm sees the farmer, it gets fed; thus, the 'turkey scientist' formulates this as a law—until Thanksgiving arrives, and the farmer shows up not with food, but with a butcher’s knife.
Hume argued that what people commonly regard as 'causality' is merely a psychological illusion; in reality, no such necessary connection exists. All we observe is that event A precedes event B, but we can never prove that A causes B.
Inductive reasoning presupposes 'causal determinism'—the belief that whenever an event occurs, there must be a cause behind it. If causality itself does not exist, then induction becomes a form of circular reasoning, which is precisely why inductive conclusions are prone to error.
Nevertheless, although Hume identified the problem with induction, he remained an empiricist, as he distinguished two types of 'credible' knowledge:
One type consists of directly observable empirical experiences, such as data and phenomena;
the other comprises purely conceptual knowledge independent of experience, such as certain mathematical truths.
The term "credible" is placed in quotation marks because Hume wanted people to be extremely cautious about their own experiences; otherwise, they might inadvertently introduce false knowledge that falls into the trap of 'causality.'
Hume’s two categories of "credible" knowledge may appear to say nothing at all, yet they represent the most fundamental truths of value investing.
5. Buffett: The Great Skeptic
Nothing embodies empiricism more than Buffett’s concept of the "circle of competence." Buffett said, "I like businesses I can understand. I start by asking whether I can understand them—using this criterion alone, I filter out 90% of companies. There’s a lot I don’t understand, but fortunately, what I do understand is sufficient."
Buffett places great importance on corporate management capability, yet he has rarely articulated any formal management theories. In his view, investing means continuously observing the actual performance of businesses, markets, and management teams, trusting common sense, and calculating intrinsic value using the simplest mathematical methods.
Buffett said, "Companies like Oracle, Lotus, and Microsoft—I simply can’t predict what their moats will look like ten years from now. Gates is the most brilliant business mind I’ve ever encountered, and Microsoft holds a massive competitive advantage, but I truly don’t know what Microsoft will look like a decade hence, nor can I precisely foresee how its competitors will evolve over that period. However, I do know what the chewing gum business will look like ten years from now. No matter how the internet evolves, it won’t change our habit of chewing gum—it seems nothing can alter that habit."
Buffett acknowledges the internet revolution and the profound changes brought by AI, but he refrains from investing in them because their outcomes cannot be reliably derived from experience.
Graham developed an elaborate framework for value analysis, yet Buffett operates with virtually no formal investment theory. He dislikes complex financial models and has criticized investment approaches that rely excessively on assumptions and formulas. He repeatedly cautions investors against making unfounded predictions about the future. In his view, it suffices to understand where a company’s long-term competitive advantage lies—this reflects a clear inheritance of Humean skepticism, as Buffett instinctively distrusts grand theoretical frameworks.
On this point, Munger clearly differs from Buffett. While Munger is also fundamentally an empiricist, he frequently emphasizes, 'To the man with only a hammer, every problem looks like a nail'—a statement that directly addresses Hume’s 'illusion of causality.'
Munger consistently advocates for 'multidisciplinary mental models,' urging the extraction of universal principles from fields such as mathematics, physics, biology, psychology, and economics to understand the real world. He promotes rational integration and deductive reasoning to guide decision-making—a distinctly 'rationalist' approach that Buffett has never favored.
Thus, Munger's investments have been bolder than Buffett's, often featuring aggressive cases that break from his personal experience to challenge the boundaries of his own investing. Munger visited China only once, in 2010, when he evaluated and invested in BYD. His later investments in Alibaba and other Chinese equities were more rooted in certain investment principles—effectively 'betraying' the traditional empiricist approach—and Buffett did not endorse them.
A better reflection of Hume’s skeptical tradition is the 'margin of safety' principle followed by all value investors. This concept is premised on the idea that anyone can be wrong—and being wrong entails financial loss. The solution is not to construct an infallible system but, as Buffett put it, to 'buy a dollar’s worth of value for fifty cents.'
Buffett once held a significant position in IBM and repeatedly praised the company at shareholder meetings, yet seven years later he ultimately admitted he had been mistaken. However, Buffett did not lose money on his IBM investment, precisely because he bought it cheaply: his acquisition cost in 2011 was around RMB 170 per share, equivalent to just 13 times earnings. After holding the stock for six years, he exited at approximately RMB 150 per share; factoring in dividends received over those six years, the investment essentially broke even.
It is relatively easy to identify whether a stock is cheap, but far more difficult to accurately assess a company’s intrinsic value. Buffett possesses exceptional acuity in judging business value and deep insight into economic and societal trends—yet he remains cautious about his own judgments, always asking first: 'What if I’m wrong?'
Throughout his life, Buffett has viewed the world—which he has observed for nearly a century—with a skeptical eye, always ready to walk away at a moment’s notice.
6. From Empiricism to Bayesianism
Finance professionals who have undergone more than 16 years of classical education are instinctively drawn to 'rationalism' in their early investment careers, always striving to build an all-encompassing investment framework.
Non-professional retail investors, on the other hand, often fall into the trap of 'empiricism,' relying on fragmented knowledge and anecdotal experience to guide their trading decisions, ultimately becoming confused about what they did right or wrong.
Overall, however, empiricism better suits the demands of high-risk activities like investing.
Epistemology constitutes only a small part of philosophical inquiry, and 'research' is merely one component of investing. Rationalist investors hold a stubborn belief that one must fully understand all underlying laws to make money, whereas empiricists—especially skeptics—adopt a more pragmatic mindset:
You don’t need to understand how a television works; you only need to know that it can show entertaining programs. Similarly, you don’t need to know all the facts—only the subset of knowledge (your circle of competence) that reliably enables you to make money.
An empiricist must also build a 'system for assessing cognitive value,' but this system is not intended to explain markets, stock prices, or company valuations. Rather, it is designed to guide action: a decision-making mechanism grounded in experience that is probabilistically sound, coupled with an exit mechanism for recognizing and correcting errors.
As Hume stated, 'A wise person adjusts their beliefs according to the evidence'—indeed, this is also the origin of Bayesian thinking.
Editor/Jayden