Pragmatic Theory of Truth

A theory of truth originating with William James and Charles Sanders Peirce, applied to investing most fully by Bill Miller. Its core claim: the truth of an idea is determined by its cash value — whether it works, whether it produces reliable predictions, whether acting on it generates good outcomes. Truth is not correspondence to a fixed, external reality; it is a property of ideas that successfully guide action.

The contrasting position, which Miller calls the correspondence theory of truth, holds that the world works a certain way, the investor’s job is to understand that structure, and investment decisions that violate it are errors. For classical value investing, the correspondence position crystallised into doctrinal form: value is defined by low P/E, low P/book, low P/revenue; anything outside those screens is speculation.

The pragmatic investor’s operating method

The pragmatist investor does not ask “does this conform to value doctrine?” but “where is value actually working, and why?” This produces two operating habits:

1. Following value migration: “Value is always in the marketplace. It just migrates.” Sectors, asset classes, and business models that are generating genuine economic returns are where value is working at any given time. The correspondence theorist is frozen in one sector (low P/E industrials, say); the pragmatist is sector-agnostic, always asking where the value is this cycle.

2. Updating descriptions: the pragmatist tests their description of a business against what is actually happening. If the description stops predicting correctly, it is discarded for one that does. This is the pragmatic theory in its dynamic form — truth is not a permanent property of a description but a property the description earns by working.

Wittgenstein’s description/explanation mechanism

Bill Miller connected the pragmatic theory to Wittgenstein’s philosophy of language in his applied investment teaching. The chain is: language → description → explanation → action. The description you choose for a phenomenon determines the explanation you construct; the explanation determines what you do.

The canonical case is Amazon (as reconstructed by Robert Hagstrom):

Miller’s insight was that most of Wall Street had the wrong description — and the wrong description is causally upstream of the wrong explanation and the wrong action. The analytical work of investing, under this framework, is primarily description work, not calculation work.

William James’s blindness diagnosis

Miller used James’s 1898 essay “On a Certain Blindness in Human Beings” to diagnose why otherwise intelligent investors persistently had the wrong description of Amazon. James visits a mountain cabin in North Carolina and calls it hideous squalor; a local mountaineer corrects him: the cabin is a triumph of family, duty, and human will carved from wilderness. James’s blindness was structural — his Cambridge academic frame literally prevented him from seeing the ideality available to someone with a different frame.

Miller’s application: the fund managers who attacked the Amazon position had the equivalent structural blindness. Their frames (value vs. growth; P/E as the definitive measure) made them constitutively unable to see what Miller was seeing. The pragmatist’s discipline is to continuously test whether their frame has produced this kind of blindness — to ask whether a better description exists that would make the current facts interpretable differently.

The 2008 lesson: description errors about government

Legg Mason’s catastrophic 2008 losses illustrate the framework’s limits. The team had successfully bet in 1992 (the S&L crisis) that the government would let financial firm equity survive — they described 2008 as analogous to 1992. The correct description was different: 2008 was a situation where committee politics (“not lending to fat cats on Wall Street”) would wipe out all equity regardless of the financial logic of preservation.

Hagstrom’s rule derived: whenever government is involved in a binary survival-or-destruction decision for a company, the outcome belongs in the “too hard” pile — political committee dynamics are not modellable by financial analysis alone. The description “this is a government-dependent binary” automatically triggers a pass.

Connection to value migration

The operational phrase that captures the pragmatic theory in one sentence is Miller’s: “Value is always in the marketplace. It just migrates.” The correspondence theorist defines value narrowly and waits for the market to bring value to their category. The pragmatist defines value by outcomes and follows it wherever it migrates.

This is the philosophical justification for Miller’s “core” portfolio that held both value stocks and growth stocks simultaneously — a portfolio that confounded consultants who wanted to put him in a style box. In Miller’s frame, there were no value or growth stocks; there were mispriced stocks and fairly priced stocks. Where the mispricing was — in low P/E industrials or high-multiple technology companies — was an empirical question, not a doctrinal one.

Where mainstream views differ

The correspondence theory defence: Orthodox value investors (Tweedy Browne, many Graham disciples) hold that the discipline of restricting to quantitatively cheap securities provides a systematic margin of safety that more than compensates for missed compounders. The constraint is a feature, not a bug: it prevents the investor from rationalising any expensive stock as having “the right description.” The pragmatist risks using flexible description as a cover for speculative purchases.

The efficient market counterposition: If markets are highly efficient, no description of a company — pragmatist or otherwise — can systematically generate alpha, because all known information is already priced. Miller’s Amazon return is, on the EMH view, a lucky combination of insight and survivorship bias, not a reproducible methodology.

The anchoring problem: Most investors who believed they were pragmatists in the late 1990s were describing speculative stocks in growth-story language that worked temporarily and then stopped. The pragmatic test (“is this description working?”) is a lagging indicator — you only learn the description was wrong after the drawdown. Correspondence-theory investors, precisely because they are inflexible, are protected from this particular failure mode.

Related

• Value Investing — the correspondence theory’s dominant form in finance; pragmatism as revision
• Knightian Uncertainty — Miller’s epistemological basis for holding concentrated positions
• Bill Miller on Amazon, Bitcoin, and Buying at a Discount to Future Value — Miller’s account
• Robert Hagstrom on Pragmatic Truth, Multi-Disciplinary Investing, and the Concentrated Portfolio — applied reconstruction
• Compounding — Miller’s Amazon holding as the worked example; future-value as the pragmatic unit
• Roots and Branches — Shayegh’s parallel framework: qualitative roots as the upstream causal factor, branches as the lagging quantitative indicators