What Five Philosophers Agreed On: Knowledge Cannot Be Made Fully Explicit

There is a recurring pattern in twentieth-century epistemology that deserves more attention than it receives. Between 1949 and 1969, five thinkers working in entirely separate traditions — philosophy of mind, physical chemistry, history of science, analytic epistemology, and ordinary language philosophy — independently arrived at an isomorphic conclusion. They each discovered that the ideal of fully explicit, articulable knowledge is not merely difficult to achieve but logically incoherent.

 

While Gilbert Ryle, Michael Polanyi, Thomas Kuhn, Edmund Gettier, and Ludwig Wittgenstein were attacking different problems using distinct methods, their findings reveal a shared structural limit to formalization. This independent convergence is not merely a coincidence of history; it is a fundamental principle of epistemology.

 

---

The Target: The Ideal of Fully Explicit Knowledge

The shared target is what Polanyi calls "the declared aim of modern science" — to establish a strictly detached, objective knowledge from which all personal, inarticulate, and unverifiable elements have been removed. In philosophy this project takes the form of the justified true belief (JTB) account of knowledge: a subject knows a proposition if and only if the proposition is true, the subject believes it, and the subject is justified in believing it. In science it takes the form of the positivist programme: knowledge is what can be explicitly stated, formally derived, and independently verified. In pedagogy it takes the form of the assumption that to transmit knowledge is to transmit rules, definitions, and propositions.

 

All five thinkers argue that this ideal, at whatever level it appears, rests on a mistake.

 

---

Ryle: The Regress of the Intellectualist Legend

Gilbert Ryle's The Concept of Mind (1949), Chapter II, "Knowing How and Knowing That" (pp. 25–58), opens the attack from philosophy of mind.

 

Ryle distinguishes two kinds of knowledge: knowing that (propositional — facts, rules, definitions) and knowing how (practical competence — playing chess, diagnosing a patient, speaking grammatically). The tempting view — which Ryle calls the intellectualist legend — is that knowing how is just knowing that applied. To act intelligently, you must first consult a mental set of rules governing the action; the intelligence of the act is derived from the prior intellectual acknowledgment of those rules.

 

Ryle's refutation is a regress argument (pp. 29–32). If every intelligent act must be preceded by consulting a rule, then consulting a rule is itself an intelligent act — and must therefore be preceded by consulting a rule for rule-consulting, and so on without end. The chain never bottoms out. The conclusion at p. 32 is precise: intelligent cannot be defined in terms of intellectual; knowing how cannot be derived from knowing that. Skilled performance is one operation, not two. The intelligence is in the doing, not in a shadow performance preceding it.

 

This means that a large and central class of knowledge — practical competence, expertise, skill — is irreducibly non-propositional. It cannot be extracted from practice and deposited into rules without destroying what it was.

 

---

Polanyi: The Structure of Tacit Knowing

Michael Polanyi's The Tacit Dimension (1966) provides the most systematic positive account of what this non-propositional knowledge actually is and how it works.

 

His central claim: "We can know more than we can tell" (p. 4). We recognise a person's face among thousands yet cannot specify the features by which we do so. An expert diagnostician detects a pattern before being able to articulate what the pattern is. A scientist senses that a problem is real and worth pursuing before being able to state why. In all these cases, knowledge is operating at a level below what can be put into words.

 

Polanyi identifies the structure underlying these cases. Tacit knowing always involves two terms: a proximal term (the particulars we attend from) and a distal term (the entity we attend to). When a blind man uses a stick to feel his way, he attends from the impact on his palm to the tip of the stick touching objects at a distance (p. 12). The impact on the palm is known tacitly — it is the medium of knowing, not its object. The structure is from-to: attending from subsidiaries toward a focal meaning.

 

This structure has four interlocking aspects:

 

• Functional: the proximal term is relied upon, not examined — it serves as the medium of attending to something else
• Phenomenal: we are aware of the proximal in terms of the appearance of the distal
• Semantic: the proximal means the distal; meaning is always displaced away from the knower
• Ontological: tacit knowing is a knowing of the comprehensive entity constituted by its two terms

 

The practical consequence is severe. To attend focally to the proximal term — to examine what one was previously attending from — is to destroy the comprehensive entity it constituted. The pianist who focuses on finger placement loses the music (pp. 18–19). This is what Polanyi calls the hazard of unbridled lucidity: the attempt to make everything explicit dissolves the coherence it was trying to examine.

 

Applied to the philosophy of science, the implications are radical. To identify what you want to formalise — a frog, a mechanism, an economic relationship — you must already know it tacitly. The formalisation depends on prior tacit knowing and can function only within a further act of tacit knowing. The ideal of a fully explicit theory that eliminates all tacit residue is therefore self-defeating (pp. 20–21).

 

Polanyi's resolution of the Meno paradox deserves specific attention (pp. 21–22). Plato showed that you cannot search for the solution to a problem because either you already know it (no problem) or you don't know what you're looking for (no search). Tacit knowing resolves this: the scientist senses the hidden coherence of a problem without being able to state it. This inarticulate foreknowledge is what guides inquiry (pp. 23–24). Every discovery arrives fraught with further indeterminate implications the discoverer cannot specify — yet commits to.

 

---

Kuhn: Paradigms and the Transmission of Tacit Knowledge

Thomas Kuhn's The Structure of Scientific Revolutions (1962) takes the same insight and applies it to the sociology of science — how scientific communities form, maintain themselves, and change.

 

In Chapter V, "The Priority of Paradigms," Kuhn argues that the coherence of a scientific tradition is not maintained by explicit rules. Scientists within a tradition agree — in their judgments of plausibility, their choice of problems, their standards of solution — without being able to state the principles underwriting that agreement. The search for explicit rules governing a mature scientific tradition repeatedly fails: every formulation turns out either too strong or too weak. In note 1 of that chapter, Kuhn names the concept directly, citing Polanyi: scientists are guided by tacit knowledge acquired through practice and that cannot be articulated explicitly.

 

The mechanism Kuhn identifies is the exemplar. Scientists do not learn their discipline by first acquiring abstract principles and then applying them. They learn through worked examples — canonical problem-solutions in textbooks and laboratory exercises — which train a perceptual capacity. After sufficient exposure to exemplars, a scientist does not apply a rule to recognise that a new problem resembles a solved one; the resemblance is seen directly, without conscious inference. What the exemplars transmit cannot be extracted from them and restated as rules without loss.

 

By the 1969 Postscript, section 3, "Paradigms as Shared Examples," Kuhn identifies this as the core of his argument: the paradigm as shared example is the central element of what he takes to be the most novel and least understood aspect of the book. The knowledge that makes normal science possible — the knowledge allowing scientists to recognise good problems, plausible results, and acceptable solutions — is not the knowledge written in papers and textbooks. It is the tacit residue of exemplary practice, transmitted through apprenticeship and reproducible only by doing.

 

---

Gettier: The Insufficiency of Explicit Conditions

Edmund Gettier's three-page paper "Is Justified True Belief Knowledge?" (Analysis, 23(6), 1963, pp. 121–123) attacks the same ideal from within analytic epistemology, through pure logical counterexample.

 

The classical account defines knowledge as justified true belief: a subject S knows that p if and only if p is true, S believes p, and S's belief is justified. Gettier produces cases showing that all three conditions can be satisfied and knowledge still be absent. His Case II: Smith has strong evidence that Jones owns a Ford, and constructs the belief "Jones owns a Ford, or Brown is in Barcelona." Jones turns out not to own a Ford, but Brown really is in Barcelona by sheer coincidence. Smith's belief is true, believed, and justified — and is clearly not knowledge.

 

The post-Gettier programme spent decades attempting repairs: add a condition requiring no false intermediate premises; require reliability of the belief-forming process; require sensitivity to the truth; require safety from error in nearby possible worlds. None has succeeded.

 

The deeper lesson — which connects to Polanyi and Ryle — is that what is missing in Gettier cases is not a further articulable condition. It is something that cannot be stated as a condition at all: the subject's evidence does not genuinely track the truth — it happens to coincide with it. The from-to relation has misfired. Polanyi's framework makes the diagnosis precise: justification is a propositional matter, but genuine tracking of reality through one's clues is prior to justification and tacit. You cannot patch the gap by adding more words to the definition because the gap is not in the definition.

 

---

Wittgenstein: Rules, Hinges, and the Bedrock of Practice

The later Wittgenstein approaches the same problem from two directions: the philosophy of language and the structure of epistemic justification.

 

In Philosophical Investigations (§§185–242), Wittgenstein works through the rule-following argument. He gives the example of a student taught the series "add two," who executes it correctly up to 1000, then writes 1004, 1008, 1012. On what grounds can we say he is wrong? Nothing in the explicit formulation of the rule determines what "add two" means beyond 1000. Any rule requires interpretation; any interpretation is itself a rule requiring a further interpretation; the regress is unavoidable. The conclusion is that rule-following is not grounded in an inner mental act of explicit interpretation but in a practice — a trained, shared way of going on. At the base: this is how we act. Not: this is what the rule says, which we have now interpreted.

 

In On Certainty (§§341–343), Wittgenstein extends this to the structure of inquiry itself. Certain propositions must remain exempt from doubt for any inquiry to function. These are the hinges on which questioning turns: the questions we raise and our doubts depend on the fact that some propositions are exempt from doubt (OC §341). If we want the door to turn, the hinges must stay put (OC §343).

 

Hinge propositions are not known in the ordinary sense. They cannot be verified by inquiry, because inquiry presupposes them. They are not held consciously or stated explicitly; they are acted from, not reflected upon. The attempt to submit them to scrutiny does not deepen knowledge — it collapses the framework within which knowledge is possible.

 

The alignment with Polanyi's subsidiaries is exact. Hinge propositions function as the proximal term at the level of the entire epistemic system: they are what we attend from, the silent background against which all focal knowing operates. Wittgenstein arrives at this through analysis of language use; Polanyi arrives at it through the structure of perception and skill. The result is the same.

 

---

The Convergence and Its Significance

The five arguments operate at different levels — logical, phenomenological, sociological, epistemological, linguistic — but share a common structure. Each identifies a domain of knowledge (practical competence, tacit foreknowledge, exemplar-based training, evidential tracking, hinge certainty) that is functionally indispensable to the explicit knowledge we prize, yet irreducibly resistant to full explicit articulation. And each shows that attempts to eliminate this residue by converting it into explicit form either fail, generate a regress, or destroy the competence they were trying to capture.

 

The practical implications extend beyond professional philosophy.

 

For scientific methodology: Ryle's regress and Kuhn's exemplars both imply that no methodology can be written down that would produce good science if followed mechanically. The tacit judgment of the experienced practitioner is not replaceable by explicit rules, and research training that treats methods as algorithms will produce sophisticated procedure-followers rather than scientists.

 

For epistemology: Gettier's cases suggest that the search for necessary and sufficient propositional conditions for knowledge is misconceived at the root. The question is not what conditions to add but what kind of thing knowledge is — and Polanyi's account suggests it is an act of attending through rather than a propositional state satisfying articulable criteria.

 

For any discipline that transmits expert knowledge: Kuhn's exemplars imply that what is most important in a training programme is not the explicit content but the worked examples through which tacit perceptual competence is built. The explicit content can be stated in a textbook; the tacit competence can only be acquired by doing.

 

What makes the convergence striking is its independence. Ryle was a philosopher at Oxford attacking Cartesian dualism. Polanyi was a physical chemist in Manchester reflecting on the conditions for free scientific inquiry. Kuhn was a physicist turned historian of science at Harvard. Gettier was an analytic philosopher at Wayne State University. Wittgenstein was doing something that defies disciplinary classification. None was primarily responding to the others. They were following their own problems to the same conclusion.

 

That conclusion, stated plainly: there is no version of knowledge — scientific, practical, perceptual, or epistemic — that does not rest on a tacit foundation that cannot in principle be made fully explicit. The explicit is always the surface of something deeper that it cannot contain.

 

---

References

 

Ryle, G. (1949). The Concept of Mind. Hutchinson. Chapter II, "Knowing How and Knowing That," pp. 25–58.

 

Polanyi, M. (1966). The Tacit Dimension. University of Chicago Press.

 

Kuhn, T.S. (1962). The Structure of Scientific Revolutions. University of Chicago Press. Chapter V, "The Priority of Paradigms," note 1; Postscript—1969, §3 "Paradigms as Shared Examples."

 

Gettier, E.L. (1963). Is Justified True Belief Knowledge? Analysis, 23(6), pp. 121–123.

 

Wittgenstein, L. (1953). Philosophical Investigations. Blackwell. §§185–242.

 

Wittgenstein, L. (1969). On Certainty. Blackwell. §§341–343.