Reason: Tacit Things We Cannot Fully Explain
I come back to material on the tacit dimension, which informed my thinking decades ago, and has never been completely absent since.
Introduction
This discussion returns to a long-standing puzzle at the heart of human cognition: how we know and do things that we cannot fully explain. It draws primarily on the concept of tacit knowledge, as articulated by Michael Polanyi, and situates it against the background of contemporary debates in philosophy, linguistics, and artificial intelligence. Polanyi’s central claim—that we “know more than we can tell”—remains not only conceptually fertile, but increasingly relevant in an age that overvalues explicitness, systematization, and formal modeling.
The exploration proceeds across domains: speech, writing, mimicry, social behavior, perception, and the emergent functions of computational models. The core theme is that much of what appears as rational, intentional, or articulate is built upon a foundation of unspoken, often unrecognized pattern recognition and practical know-how. Language is treated not as the basis of thought, but as a secondary process—an output of deeper, less accessible operations.
Philosophers, linguists, and AI theorists have often misunderstood or underappreciated this dimension. In mistaking language for thought, or rules for understanding, they have overlooked the real structure of knowledge as lived, enacted, and embodied. This discussion reasserts the primacy of the tacit dimension as a necessary corrective to reductive accounts of reason, meaning, and cognition.
Discussion
We Can Do Far More Than We Can Explain
Most of what human beings do—perceiving, moving, choosing, creating—we cannot fully explain. We carry out actions fluently, often without knowing how we do them. We connect events, name things, form judgments, produce language, respond to music, recognize faces, and mimic behaviors—all without explicit awareness of the internal processes involved. Language itself, with its shifting boundaries, multiple meanings, and fuzzy categories, is more often used after the fact to make sense of what has already occurred. Explanations are often retrospective scaffolding—sometimes accurate, often not. Even when causality is presumed, our ability to assign causes to most of what happens within us, or around us, is limited. This reaches well beyond the domain of formal knowledge and enters the realm of what Michael Polanyi called the tacit—a dimension that resists full articulation but structures virtually all cognition and experience.
An Overview of Michael Polanyi’s Views
Michael Polanyi’s work presents a comprehensive response to the mistaken belief that knowledge must be explicit to be real. His philosophy emerged as a critique of the dominant objectivist conception of science and rationality. He argued that “we know more than we can tell,” and insisted that not only science, but all meaningful human activity, rests on pre-articulate, embodied forms of knowing. This tacit dimension is not vague or irrational—it is structured, skilled, and often more reliable than formal reasoning. His major works—Science, Faith and Society (1946), Personal Knowledge (1958), and The Tacit Dimension (1966)—systematically dismantle the illusion that knowing is reducible to impersonal statements or mechanical procedures.
1. Principal Works on Tacit Knowledge
Science, Faith and Society (1946): Challenges scientific objectivism; introduces the role of personal judgment in discovery.
Personal Knowledge (1958): Develops the idea that all knowledge is grounded in personal involvement and commitment; a defense of the post-critical stance.
The Tacit Dimension (1966): Outlines the structure of tacit knowing as a movement from subsidiary to focal awareness; accessible synthesis of his larger epistemological vision.
2. Core Concepts in Polanyi’s Theory
a. Tacit Knowledge:
Tacit knowledge includes skilled actions and recognitions—riding a bicycle, crafting a musical phrase, interpreting a gesture—that cannot be reduced to formulas. It is structured and directed, but not fully verbalizable.
b. The From–To Structure of Knowing:
All knowing is from subsidiary awareness to focal integration. One sees a pattern in an x-ray, hears a phrase in music, or interprets an expression not by assembling explicit rules but by integrating clues into a coherent whole.
c. Personal Knowledge:
Knowledge is always personal in the sense that it involves belief, trust, and commitment to a reality that transcends the knower. Even in science, these fiduciary acts are foundational, not dispensable.
d. Indwelling:
To “indwell” a set of skills or symbols is to internalize them so deeply that they become extensions of perception and thought—like the blind man’s cane or the musician’s instrument.
3. Implications Across Domains
Polanyi’s ideas have broad implications:
Science and Discovery: Discovery is guided by intuition, heuristics, and unformalizable pattern recognition—not algorithmic deduction.
Education and Apprenticeship: Learning is not about transferring facts but about participating in forms of life—absorbing ways of seeing, doing, and being.
Artificial Intelligence: Symbolic AI fails to model human intelligence because it overlooks the embodied, skillful, and non-explicit nature of most human knowledge.
Philosophy of Mind: Polanyi’s account challenges the idea that mind is reducible to information processing. Knowing is lived, not just computed.
4. Legacy and Reception
Although often marginalized in analytic circles, Polanyi's influence has been considerable:
Kuhn’s paradigms, Collins’ tacit knowledge in science, and post-cognitivist philosophy of mind build on his insights.
Professional training, craft learning, and embodied cognition all draw from Polanyi’s claim that knowledge is enacted and shared, not merely encoded and transmitted.
His work continues to resonate in epistemology, education, cognitive science, and practice-based theory.
Cognitive Processing Occurs Under the Covers
Polanyi’s framework clarifies why so much cognitive activity remains unspoken, unnoticed, and untraceable. Mental operations occur beneath conscious inspection. Thought arises, unbidden and unstructured, until language retrospectively tries to fix it. For any one idea, dozens of phrasings are possible. Words give shape to inchoate mental states, but they never fully capture them. This is not just a quirk of communication; it is a fundamental feature of mind.
The Tacit as Diffuse Awareness
Tacit knowledge often begins as a diffuse sense of something—an unease, an impression, a recognition without words. It may be felt before it is known, seen before it is understood. With the right language, or the right metaphor, or the right analogy, it can crystallize into something communicable. But its origins lie in felt experience, not formal structure. The process of articulation is revelatory, not mechanical.
Using Words to Clarify the Tacit
Even in highly articulate domains—art, music, science, literature—what we notice is often first recognized without words. We compare patterns, sense coherence, or detect incongruity before we can explain it. Language may come later, as a tool for scaffolding thought or justifying perception. But the recognition itself precedes the explanation. Tacit knowledge makes value judgment possible—whether in aesthetics, ethics, or reasoning. And the judgment often rests on something seen but not yet said.
Mistaking Language for Thinking
This is where some philosophical traditions go astray—mistaking language for thought itself. Theorists who reduce cognition to syntax or semantics confuse the outer skin of expression with the inner structure of knowing. This is the trap of self-referential linguistic theory: taking the tools of thought to be the whole of thought. Language is derivative; it is not the source of cognition. It is what follows awareness, not what produces it.
Tacit Is Just a Placeholder
“Tacit” is a placeholder term, not a full explanation. It acknowledges that much of what we do, feel, and know operates outside conscious language. It names the shadowed regions of the mind, not to mystify them, but to respect the complexity of what cannot be reduced. To call something tacit is to say that we use it, rely on it, and trust it, even though we cannot explicate it. And that is not a weakness in human cognition—it is its foundation.
Conclusion
Michael Polanyi’s insights, and the reflections they provoke, point to a foundational truth: articulate knowledge rests on inarticulate structure. We are not machines encoding propositions—we are embodied beings who sense, act, perceive, and respond before we explain. The tacit is not a deficiency of speech but the ground of its possibility. The mystery is not that we can't always explain what we do. The mystery is that we can do it at all. And anyone who thinks this isn't a mystery hasn't thought about it.
The Mystery of the Emergence of Thought
Thought arises in the mind not as a product of deliberate manufacture but as something more like a spontaneous appearance. This fact is so familiar that it often escapes notice. Yet when examined closely, the emergence of thought reveals itself as a profound mystery—perhaps one of the deepest in all of human experience.
Thought is not identical with language, although language often follows or accompanies it. Nor is it the same as visual imagery, though some thoughts appear in pictures or imagined scenes. It is not emotion, though it may be laden with affective tone. It is not even perception, although it may be triggered by sensory input. One might try to reduce thought to awareness, but that too is insufficient, since there are states of awareness without thought—such as pain, alertness, or undirected presence.
What thought is, then, remains elusive. It may begin as a kind of inchoate stir in the background of consciousness: an itch of association, a feeling of a connection trying to surface, a phrase on the tip of the tongue. We recognize the phenomenon in expressions like, “Let me collect my thoughts,” or, “I had something to say, but I’ve lost it.” These experiences confirm that thought can exist in a pre-verbal state, felt but not yet formed, and that verbal articulation often lags behind cognitive presence.
Thoughts do not queue up neatly. They arise out of sequence, out of context, sometimes out of nowhere. A memory surfaces unbidden; a realization interrupts a mundane task; an absurd idea appears while trying to focus. Some thoughts seem random, some associative, and some coherent and goal-directed. But even in the latter case, the content of a given thought—its wording, its focus, its emotional coloration—is not consciously chosen in advance. It simply appears, as though brought forth by an internal process over which we have only partial, and mostly indirect, influence.
Does It Make Sense to Will a Thought?
To say, “I will now think a specific thought,” presumes the prior existence of that very thought—or at least some skeletal version of it—before the act of willing. But this leads to a paradox. To will a thought, one must already have the content in mind, which defeats the premise that it is being willed into existence. If one does not already know the thought, one cannot intend it. This circularity collapses the idea of volitional thought production.
The metaphor of bubbles rising in carbonated water is apt. Thoughts rise from unseen depths—sometimes relevant, sometimes irrelevant, sometimes sharp, sometimes vague. They emerge without conscious preparation, and while we can shape them after the fact, we cannot manufacture them from scratch at will. The best we can do is prepare a context in which certain types of thought are likely to arise: engaging in reflection, asking a question, setting an intention. But the thought itself must arrive of its own accord.
Neurologically, this remains a black box. Imaging studies can identify activation patterns in brain regions associated with various cognitive tasks, but this is not the same as understanding how thoughts emerge. Correlation is not explanation. To date, neuroscience has provided maps of where, and descriptions of what, but not mechanisms of how in any meaningful cognitive sense. Thought remains unpredictable in content, spontaneous in form, and largely inexplicable in origin.
The Minds of Animals and Infants
The mystery is not confined to reflective adults. Consider the minds of those who have no access to formal language:
Infants, who exhibit intention, recognition, anticipation, and response before they acquire words.
Animals, who navigate the world, form expectations, solve problems, and express emotions without ever formulating propositions.
Individuals born deaf who grow up without exposure to sign language or writing, yet still engage in reasoning, memory, and internal awareness.
In all of these cases, cognition clearly occurs. There is attention, decision-making, and emotional differentiation, but no articulate language. Thus, we must conclude that thought does not require words. Language may extend and refine thought, but it is not its precondition.
Moreover, much of what occurs even in fully articulate human minds never reaches the threshold of awareness. Reactions happen before conscious registration. Emotional shifts arise without identifiable triggers. Physical actions are taken on the basis of perceptions or evaluations that were never verbalized. All of this suggests that awareness itself is only a thin layer above a deeper and largely tacit substrate of cognitive activity.
Tacit as a Placeholder
In light of all this, the term “tacit” functions not as an explanation but as a concession of ignorance. It marks the boundary beyond which our language, our models, and our introspection cannot reach. It says: something is happening here—we can see its outputs, feel its effects—but we do not know how it works.
To call something tacit is to admit that:
We cannot specify its structure in language.
We cannot access its workings directly.
Yet we rely on it constantly, and it shows itself in our actions.
This applies not only to motor skills or social intuition, but to the basic functioning of the mind itself. Most of what we do cognitively—perceiving, prioritizing, preparing, associating, choosing—happens beneath the threshold of explanation. That is why the emergence of thought is mysterious. It is not a mystery in the sense of being magical, but in the sense of being formally intractable. We observe its manifestations, but not its genesis.
Mimicry, Linguistic Scaffolding, and the Tacit
To speak of mimicry is already to enter a field of conceptual ambiguity. Like most terms in natural language, it is polysemous, carrying multiple, overlapping meanings. Mimicry can refer to deliberate imitation, spontaneous pattern absorption, stylized replication, or even biological camouflage. But in the human cognitive and behavioral domain, mimicry generally refers to the tacit reproduction of patterns—in language, gesture, tone, phrasing, posture, or behavior—without necessarily copying their surface features exactly. What is mimicked is not the form itself, but its structure, rhythm, emotional timbre, or functional signature.
When someone mimics another’s way of speaking, they do not recreate the precise phonemes or syntax, but rather the essence—a combination of intonation, pacing, vocabulary range, and phrase construction that conveys a recognizable style. This is not mimicry as duplication. It is mimicry as patterned emulation, and it operates at a level that precedes full awareness. Most people cannot describe how they accomplish such mimicry, but they can do it. They can “sound like” someone else. And others can recognize when the imitation is good or poor—again, often without being able to say exactly why.
This form of mimicry lies deep in the tacit realm: the domain of knowing without knowing how, of doing without being able to explain. As Michael Polanyi observed, we know more than we can tell, and mimicry is one of the most immediate demonstrations of that principle. Even small children, long before they possess metalinguistic awareness, can mimic tones of voice, facial expressions, and sentence patterns. They do not know the rules—but they can perform them.
Making the Tacit Explicit
Occasionally, what was once tacit can be brought into focus and described. This process often involves linguistic scaffolding—a metaphor for the supports that allow complex understanding to take shape in language. These may include metaphors, analogies, glosses, stepwise instruction, or guided practice. For instance, a teacher may say, “Listen to where the stress falls in that sentence,” or “Try to match the rhythm of their phrasing,” helping a student become aware of elements previously enacted but unexamined.
Still, this act of scaffolding is necessarily imperfect. Language is always a poor proxy for experience. It is ambiguous, partial, imprecise, and laden with cultural and contextual assumptions. The word rhythm, for example, might gesture toward an internalized musical sense, but it cannot fully capture the felt timing or kinaesthetic flow involved in actual performance. We describe what we can, but the richness of the original experience is compressed by articulation.
Yet even with its limitations, language can sometimes illuminate the tacit, at least partially. Through careful analogy, guided observation, or extended trial-and-error reflection, individuals can draw aspects of their tacit knowledge into focal awareness. A musician may not be able to explain precisely how they shape a phrase, but after hearing a recording and discussing it with peers, they may discover vocabulary and conceptual handles that make it more accessible.
This transition—from tacit to explicit—is neither automatic nor complete. It is partial revelation, not translation.
Mimicry Has a Large Tacit Component
All human mimicry depends on this structure of subconscious pattern recognition and enactment. It does not require abstract reasoning or formal instruction. The ability to mimic accents, speaking styles, or even thought patterns emerges from exposure, perception, and internalization. It is grounded in recognition without propositional representation. This is why mimicry often appears as a gift, or “natural talent”—some individuals seem to pick up voices, gestures, or musical styles with uncanny accuracy, yet cannot explain how.
But in reality, this is not mystical; it is tacit cognitive ability at work. The brain recognizes structural regularities—timing, pitch, phrase contour, tone modulation, syntactic shape—and reproduces them through motor and vocal output, often below the threshold of conscious processing.
Speech itself may be the paradigmatic case. Children learn to speak by hearing speech. They do not consciously learn syntax, grammar, or phonetics. They absorb the rhythms, intonations, and semantic patterns of the surrounding language and begin to produce speech that conforms to the heard model. This is mimicry, not memorization. And it occurs with minimal reflection, which is why most speakers are unaware of the structure of their own language. They “just speak.” Linguists and grammarians come later—after the tacit processes have done their work.
And the same applies to the adoption of dialects, idioms, and speech registers. A person may, without planning or awareness, begin to speak differently depending on social context—a different vocabulary here, a change in cadence there. This too is mimicry, filtered through social immersion and tacit adaptation. It is rarely reflected upon, but easily recognized by others.
Conclusion
Mimicry, like language itself, rests on tacit capacities that operate beneath conscious awareness, and only occasionally surface into verbal form. When we mimic, we emulate the form of the form—the recognizable shape of behavior or expression—not by rule-following, but by perceptual pattern matching and embodied rehearsal. And though we may describe parts of it—through linguistic scaffolding, metaphor, or analogy—we never fully capture the thing itself in language.
The phenomenon thus spans a continuum: from purely tacit reproduction to partially scaffolded awareness to formal analysis. But even the most technical descriptions rest on the prior existence of a non-explicit grasp of the pattern. That grasp is learned, enacted, and rarely articulated. It is felt before it is known, and known before it is said. And anyone who thinks this isn't a mystery simply hasn’t thought about it.
Mimicry and the Tacit: An Expanded Inquiry into Unspoken Understanding
The material presented sketches an implicit theory of cognition, skill, and social behavior grounded in mimicry and tacit knowledge—a combination that governs far more of human life than most explicit theories can accommodate. What follows is a fully elaborated exploration of those themes across domains, framed by the recognition that human intelligence, communication, and cultural participation depend on the internalization of patterns that are not, and perhaps cannot be, fully articulated. That we can perform such mimicry—fluently, precisely, responsively—without knowing how, and often without knowing that we are doing it, signals a fundamental limitation in any model of mind that privileges symbolic reasoning or conscious intent.
Mimicry of Speech: The Paradox of Imitative Fluency
Among the most remarkable forms of mimicry is the ability of certain individuals to reproduce the vocal patterns, lexical choices, and rhetorical cadences of others with striking accuracy. Imitators of public figures—actors, comedians, political satirists—often evoke laughter or recognition not because they say the same things, but because they capture the essence of how someone sounds and speaks. This includes tone, tempo, stress, phrasing, pitch variation, and often subtle characteristics such as the rhythm of pauses or the idiosyncratic turn of a phrase.
This capacity seems to be partially innate, partially learned through exposure, and heavily tacit in its execution. Skilled mimics cannot usually explain how they do it, and unskilled mimics cannot simply be trained through instruction alone. The process appears to rely on an internal recognition of structural pattern, an ability to perceive what is distinctive, and a motor system capable of reproducing it without conceptual breakdown. It’s imitation without replication, and learning without conscious formulation.
Mimicry of Dialect and Linguistic Context
Language use is shaped by context, identity, and immersion. People unconsciously absorb dialectical features—phonetic variations, vocabulary shifts, idiomatic constructions—simply by being in a speech environment. A person who moves to Appalachia and later emerges speaking with an Appalachian inflection has not studied a rulebook. They have absorbed and emulated—often to their own surprise.
This is no different from children acquiring language. They do not deduce grammar. They do not diagram syntax. They mimic speech, map forms onto use, and adapt to the style of their social group. The same applies to intellectual dialects: an undergraduate studying philosophy may begin to “write like a philosopher”—with recursive qualifiers, modal hedging, and abstract nominalizations—not through design, but through exposure to a model of discourse that becomes second nature.
This acquisition is not just about vocabulary or sentence structure. It involves adopting a style of thought, an orientation toward meaning, and a set of inferential habits—again, typically without being able to articulate how they were acquired. The process is osmotic, not propositional.
Mimicry of Written Style: Reproducing Textual Identity
Just as vocal style can be mimicked, so too can written expression. Some individuals possess an intuitive ability to reproduce the stylistic fingerprint of another author—word choice, sentence rhythm, tone, use of metaphor, density of abstraction, and even grammatical quirks. This ability, like vocal mimicry, depends on pattern recognition, but the patterns are encoded visually and cognitively rather than aurally.
The mimicry of style does not require semantic copying. What is mimicked is form, not content. A skilled parodist can produce prose that “feels like” Hemingway or Le Carré even when saying something utterly foreign to their interests. This too, however, defies reduction. One cannot easily list the rules that make one’s writing feel “like” someone else’s. The recognition and reproduction of style depend on a felt sense of congruence—not a formula.
This parallel between human mimicry of style and AI-generated stylistic output suggests a convergence: both systems work through statistical approximation, emulating structure without necessarily grasping content. Whether the human brain mimics algorithmic computation, or whether computation is merely a rough analogy for the brain’s own statistical learning, remains an open question. But the capacity to recreate the essence of form from distributed patterns is a shared characteristic of both.
The Tacit and Bodily Mimicry
Mimicry is not limited to voice or text. Many individuals can emulate body language, facial expressions, and gesture with astonishing fidelity. Actors, impersonators, dancers, and certain athletes develop acute sensitivity to kinetic pattern and postural nuance. They learn how someone moves—how they smile, pause, shift their weight, raise their eyebrows—not by description, but by seeing and doing.
What they mimic is again the essence: the recognizable signature of movement, which conveys personality, mood, or identity. And it is just as often tacit. Ask someone how they mimic a walk or a stance, and they may struggle to explain. Ask an audience whether the mimicry is good, and they will recognize it instantly.
This capacity is innately human and emerges early. Children mimic adults, not just in speech but in posture, gaze, rhythm, and emotional display. The mechanism appears to be embodied simulation—we perceive others by internally recreating their movements, an ability likely grounded in the mirror neuron system, but still poorly understood.
Mimicry of Social Behavior: Osmosis and Aspiration
Perhaps the most subtle—and socially consequential—form of mimicry occurs in the adoption of class behaviors, status signals, and group norms. People often begin to speak, act, and present themselves in ways consistent with the social group they inhabit or aspire to join. This includes speech register, mannerisms, grooming standards, emotional display norms, and even moral rhetoric.
This process is rarely taught explicitly. It is almost always acquired tacitly, through immersion, exposure, and perceptual sensitivity to social cues. One learns how to “sound educated,” “act corporate,” or “appear streetwise” not by reading a manual, but by noticing what fits and what doesn’t, often below conscious awareness.
The same principle governs detectability: observers are quick to spot when someone’s behavior is off-key for their professed social role. Small mismatches in tone, vocabulary, timing, or gesture can betray inauthenticity. This ability to detect discordance is itself a tacit skill, and again, no one can explain precisely how it works. They “just know.”
Social Mimicry in Conversation: Adaptive Fluency
Ordinary conversation is a fluid, constantly adjusted performance. Speakers adapt their language, tone, and content depending on their perception of the listener. They shift register, simplify or complexify vocabulary, modulate formality, and tune their pacing—all in real time. Most of this occurs below the level of conscious control.
This too is mimicry. We match the social style of others—aligning ourselves, reducing friction, facilitating connection. The goal may be empathy, persuasion, or just social ease. But the mechanism is the same: a largely tacit tuning of behavior to fit context. When this goes wrong—when we “talk down” to someone or “miss the tone”—we feel it immediately. Yet we cannot always specify what the correct tone would have been.
Facial Recognition and Tacit Perception
Facial recognition may be one of the most robust forms of tacit pattern recognition. Most people can instantly recognize thousands of faces, even with poor lighting, partial occlusion, or years of separation. But few, if any, can say what they are recognizing. It is not simply eye spacing or jawline curvature. It is a gestalt, a total configuration.
This capacity can be lost through neurological insult (e.g., prosopagnosia), confirming its deep biological substrate. But when intact, it operates without conscious decoding. One sees a face and knows. The same applies to perceived racial characteristics, family resemblances, or emotional states—each recognized quickly, unreflectively, and often more accurately than verbal description could convey.
Musical Mimicry and the Tacit Grammar of Sound
Musical creativity, too, is grounded in mimicry. New melodies emerge from internalized structures, accumulated from listening and experience. Musicians do not create from nothing. They create from fragments of heard motifs, harmonic frameworks, rhythmic tendencies—recombined and transformed. Even those who claim to be wholly original work within tacitly absorbed idioms.
An improviser may not “know” consciously which notes they’ll play next, but their hands find them—based on thousands of prior patterns absorbed into motor memory. This is not rule-based composition; it is tacit navigation of a structured sound space. And the distinction between the talented and the mediocre may lie not in rule-following capacity but in intuitive structural fluency—a deeply tacit sensitivity to what fits.
Recap: Tacit Understanding and Mimicry as a Cognitive Foundation
All of the above points to a single, unified theme: mimicry is the behavioral expression of tacit understanding, and tacit understanding is the cognitive infrastructure upon which culture, communication, and coordination are built. This is not marginal. It is foundational.
From conversation to composition, from gesture to grammar, human beings act out patterns they cannot explain, respond to structures they cannot define, and create variations on forms they do not explicitly know. This is not failure. It is the default condition of embodied cognition.
Further Domains for Exploration
Additional arenas where mimicry and tacit understanding govern behavior include:
Gestural timing and nonverbal rhythm in conversation
Turn-taking and interruption management
Humor and irony recognition
Learning of tools and crafts
Social signaling in dress and posture
Norm navigation in etiquette and ritual
Each involves pattern recognition without formulation, reproduction without propositional content, and performance guided by non-verbalized cues.
Conclusion
To reflect seriously on mimicry is to confront the limits of explicit knowledge. The human mind operates not through instruction-following, but through pattern-saturation, imitation, and contextual response. We do not deduce the world. We absorb it, internalize it, and act within its constraints. This is not an imperfection. It is the primary architecture of human intelligence. And anyone who thinks this isn’t a mystery hasn’t thought about it.
Do Philosophers Mistake Language for Thought?
A recurring and serious problem in 20th- and 21st-century philosophy—especially within the analytic tradition—is the conflation of language with thought. Philosophers too often assume that to understand thought is to analyze propositions, syntax, reference, or truth conditions. In this tradition, thinking is seen as the manipulation of linguistic tokens, whether in the head or on the page, and meaning is located within language systems, not in minds, bodies, or the world as directly experienced. This outlook can be traced back through logical positivism, Fregean semantics, early analytic philosophy, and linguistic philosophy more broadly.
But this is, at root, a category mistake. Thought and language are not identical. They are not even coextensive. To think is not necessarily to speak, write, or form propositional structures. Infants think. Animals think. Individuals deprived of language by trauma or neurological disorder still experience, infer, decide, remember, and imagine. One may have a persistent mental state or intention without ever putting it into words.
Yet within much of analytic philosophy, this is insufficiently acknowledged. Instead, the discipline has operated as a closed language game, as Wittgenstein later recognized and problematized—though even his later work can be accused of failing to escape the linguistic loop it critiques. The discipline increasingly became a self-referential exercise: philosophers analyzing the terms other philosophers use, debating the conditions under which statements are meaningful, and implicitly assuming that what is not expressible in language does not count as knowledge or even cognition.
This insularity—a community speaking to itself in self-sustaining dialects—gradually eroded contact with the psychological, phenomenological, biological, and developmental realities of human thought. What was lost was not rigor, but relevance.
Maybe Analytic Philosophy Lost Its Way a Long Time Ago
If Polanyi’s work—rich in phenomenology, cognitive insight, and epistemological realism—has been marginalized within analytic philosophy, that fact is not a critique of Polanyi, but an indictment of the discipline’s narrowing scope. His work speaks directly to real-world experience, scientific practice, and the actual structure of human knowing. That such a contribution is considered marginal indicates that analytic philosophy may have lost its bearings, having prioritized formalisms and linguistic puzzles over substantive engagement with the complexity of mind.
Polanyi’s assertion that “we know more than we can tell” cannot be reduced to a semantic ambiguity. It’s an ontological claim about the architecture of cognition. That this was not central to analytic discourse may reflect a deeper institutional drift: a tendency to substitute linguistic clarity for epistemic adequacy.
Problems with Chomsky’s Views on Language
Chomsky’s theory of language—particularly his transformational-generative grammar and the hypothesis of a universal grammar encoded in human neurology—was revolutionary in its day. But the core formalism he developed was deeply limited in scope. It treated syntax as the autonomous center of linguistic analysis and posited that semantics, pragmatics, and social context were peripheral, if not irrelevant, to the study of language.
This perspective cannot account for the vagaries, irregularities, and context-dependencies of natural language. Nor can it account for mimicry, language change, or the ways language is learned through social interaction. By treating syntax as isolated from semantics—and by assuming it must be instantiated in a uniquely human neural substrate—Chomsky’s framework effectively decontextualized language from thought, language from society, and language from communication.
Moreover, the strict separation of syntax and semantics violates not just common sense but observed linguistic phenomena. It is abundantly clear that syntax adapts to meanings, and that meanings are shaped by how syntax is deployed. Their interdependence is pragmatically obvious, even if not easily formalized.
That this model still has institutional defenders in certain quarters may say more about academic inertia than about explanatory success. Chomsky’s influence persists, but his explanatory reach has shrunk, especially as more empirically grounded theories in developmental psychology, sociolinguistics, and cognitive science have emerged.
Why Does Anybody Listen to Chomsky in the Current Age?
That Chomsky’s model continues to attract attention despite these limitations is perhaps a testament to the institutional conservatism of some sectors of linguistic theory. It also speaks to the appeal of formalism in academic disciplines: it is tidy, it generates computable models, and it offers a vision of language that is internal, rule-bound, and theoretically clean. But real language use is messy, socially embedded, and learned through exposure and interaction, not via internal rule tables or innate syntactic templates.
The notion that syntax is exclusively human, hardwired, and detached from meaning also clashes with both evolutionary logic and comparative neurobiology. If language evolved, then it must have done so through gradual elaborations of pre-existing perceptual and communicative capacities. There is no coherent reason to assume a sudden, total leap from non-syntax to syntax via a unique cognitive mutation. That assumption betrays a desire for uniqueness, not an empirical conclusion.
Computational AI and Emergence: Parallel Mysteries
The behavior of large-scale computational models, including modern AI systems, offers a revealing counterpoint. These systems, when trained on massive corpora, begin to exhibit patterned, often insightful outputs. The computations are well-understood in a technical sense—matrix multiplications, optimization functions, gradient descent—but the emergence of coherent linguistic behavior remains puzzling. It is not designed into the system; it arises from the structure and scale of exposure.
This leads to a legitimate parallel: human cognition may also rely on statistical exposure to patterns—social, linguistic, perceptual—and its emergent intelligence may be a function of non-explicit mimicry, absorption, and internalization, rather than symbol manipulation or rule retrieval. If that is so, then the strictly symbolic or rule-based models of language and thought not only fail to explain human behavior—they also fail to explain how current machines are beginning to simulate it.
Moreover, if machines can mimic reasoning without awareness, we should ask whether our own cognition is as nontransparent to itself as theirs is to us. Perhaps human thought is also emergent, not willed, and far less propositional than many philosophers have assumed.
Summary
1. The Tacit Dimension Is Foundational, Not Peripheral
Tacit knowledge undergirds most human capacities. From skilled movement to artistic expression, from social adaptation to language acquisition, what we can do precedes what we can explain. Polanyi’s “from–to” structure of knowing—the integration of subsidiary clues into focal awareness—helps clarify why formal accounts of cognition remain incomplete.
2. Mimicry Is Tacit Knowing in Action
Across speech, writing, gesture, and music, mimicry demonstrates our ability to recognize and reproduce structure without conceptual access to that structure. We don’t replicate precisely—we emulate essence, style, affect, and function. This is not imitation by rule, but performance by internalization. The tacit patterning that allows us to do this often operates below conscious awareness and cannot be fully verbalized.
3. Language Is Derivative, Not Constitutive, of Thought
Contemporary philosophy has too often conflated language with thought. But thinking precedes speaking. Cognitive life is rich in pre-verbal forms—intuitive, emotional, associative, perceptual—and to treat only propositional statements as thought is to erase most of human awareness. Chomskyan syntax models and much of analytic philosophy fail to accommodate this.
4. Emergent Computation Suggests Parallel Mysteries
AI systems now demonstrate surprising emergent behaviors. While the underlying computations are known, why these models produce fluent and contextually sensitive output remains opaque. This suggests that human cognition, too, may involve nontransparent, emergent, and non-rule-based processes, more akin to statistical internalization than to logical deduction.
5. The Tacit Refuses Final Explanation
The tacit is not a failure to know, but a different form of knowing—embodied, affective, enacted. It can sometimes be illuminated, scaffolded, or described, but it resists being codified into rules. Any attempt to reduce human knowing to what can be stated overlooks the very ground on which reason rests.
Readings List
Foundational Works by Michael Polanyi
Polanyi, M. (1946). Science, faith and society. University of Chicago Press.
Polanyi, M. (1958). Personal knowledge: Towards a post-critical philosophy. University of Chicago Press.
Polanyi, M. (1966). The tacit dimension. Doubleday & Co.
Commentary and Extensions
Grene, M. (1977). Tacit knowing and the pre-reflective self. Journal of the British Society for Phenomenology, 8(3), 164–171.
Prosch, H. (1986). Michael Polanyi: A critical exposition. State University of New York Press.
Gelwick, R. (1977). The way of discovery: An introduction to the thought of Michael Polanyi. Oxford University Press.
Cognitive Science and Embodied Mind
Clark, A. (1997). Being there: Putting brain, body, and world together again. MIT Press.
Varela, F. J., Thompson, E., & Rosch, E. (1991). The embodied mind: Cognitive science and human experience. MIT Press.
Dreyfus, H. L. (2002). Intelligence without representation—Merleau-Ponty’s critique of mental representation: The relevance of phenomenology to scientific explanation. Phenomenology and the Cognitive Sciences, 1(4), 367–383.
Language, AI, and the Limits of Explicit Models
Winograd, T., & Flores, F. (1986). Understanding computers and cognition: A new foundation for design. Ablex Publishing.
Boden, M. A. (2006). Mind as machine: A history of cognitive science (Vols. 1–2). Oxford University Press.
Gopnik, A. (2020). Infants, AI, and the future of thinking. Daedalus, 149(1), 80–94.
This material frames tacit knowledge not as a mysterious residue of otherwise clear processes, but as the principal substrate of human reason—a dimension modern thought has too often ignored, and which now must be reclaimed.