There is a moment, familiar to physicists and poets alike, when a word stops working. You reach for it the way you reach for a light switch in a power cut — the gesture is right, the word is there, but nothing illuminates. The universe has done something it was not supposed to do, and language, for a breath, cannot follow.

This essay is about that gap: the space between what we discover and what we can say — and why that gap matters not just for science communication but for thinking itself, because the words we have shape the thoughts we can reach for.

I. Language as the Shape of Thought

The idea that language does more than name a pre-existing reality — that it cuts the world at its joints, and cuts it differently depending on which language you speak — is both old and contentious. Its modern form begins with Edward Sapir and Benjamin Lee Whorf. Sapir argued in 1929 that the real world is built up, without our knowing it, on the language habits of the group (Sapir, 1929). Whorf pushed further: the grammatical system of a language is not merely an instrument for voicing ideas but the shaper of ideas, the program and guide for mental activity (Whorf, 1956). The strong version of that claim — that language determines thought outright — has not survived empirical scrutiny. The weak version has, and its implications are more unsettling than they first appear.

The experimental record is consistent across domains. Mandarin speakers, whose language makes temporal vertical metaphors habitual, are measurably faster to confirm vertical temporal relations when primed with vertical spatial images — the grammatical habit has become a cognitive resource (Boroditsky, 2001). Russian speakers, whose language compels a distinction between light and dark blues, are faster at discriminating blues that cross that linguistic boundary than those that fall within a single category; introduce verbal interference and the advantage vanishes entirely, confirming the effect is specifically linguistic and operating live during perception (Winawer et al., 2007). Speakers of languages that replace egocentric "left" and "right" with absolute cardinal directions maintain an unconscious real-time compass that English speakers never develop (Levinson, 1997). The Kuuk Thaayorre, whose spatial and temporal vocabulary is organised around cardinal orientation, arrange temporal sequences east to west regardless of which direction they are facing — following the arc of the sun, not the axis of their own body (Boroditsky & Gaby, 2010). Regier and Kay (2009) synthesised these findings and concluded that Whorf was half right: language shapes cognition reliably but partially, most powerfully in tasks that engage verbal working memory and least in purely spatial or imagistic reasoning. Language is not a prison. It is a persistent tilt in the landscape of thought.

Not everyone accepts even this weak version. Steven Pinker argued in The Language Instinct that much of Whorf's evidence was methodologically flawed, that thought proceeds in a pre-linguistic "mentalese," and that the surface structures of different languages are comparatively shallow variations on a common cognitive architecture (Pinker, 1994). Guy Deutscher, more charitably, accepted that linguistic categories influence habitual cognition while insisting the effects are modest and culturally contingent — that language shapes certain habits of mind without defining the boundaries of cognition itself (Deutscher, 2010). The response from experimental linguistics has been that this is precisely what the weak hypothesis claims: not that a language makes some thoughts impossible, but that it makes some thoughts less available, slower to reach, and less automatically recruited. Pinker's mentalese and Regier and Kay's partial effects are not incompatible — they describe different levels of the same cognitive architecture. What matters for this essay is the level at which language operates: not the deep structure of what the mind can think, but the surface structure of what it habitually reaches for. That surface, as the experimental evidence shows, is shaped by the words available.

Hold that finding against what physics has discovered. The tilt English grammar imposes on temporal cognition — the assumption of horizontal flow, of before and after, of events arranged in a line — is not a stylistic inconvenience when the object of study is quantum mechanics or cosmology. It is a structural mismatch between the representational tool and the territory. The mind Boroditsky measured is the same mind asked to conceptualise a quantum field with no definite location, a wavefunction that describes all possible outcomes simultaneously, a singularity prior to which the word "before" has no referent. The experiments on colour and time are calibration data for how badly our inherited conceptual vocabulary is likely to mislead us at the extreme frontier of physics — and how deep the misrepresentation goes.

Noam Chomsky's work complicates the picture in an important way. In his 1959 review of Skinner's Verbal Behavior and his subsequent theory of Universal Grammar, Chomsky argued that the human mind arrives pre-equipped with an innate language faculty — a biologically fixed set of structural principles shared by all human languages, on top of which surface variation is comparatively shallow (Chomsky, 1959, 1965). On this view, the causal arrow runs opposite to Whorf's: it is not that language sculpts the mind's conceptual architecture, but that an innate, universal architecture makes any language acquirable at all. This essay does not need to resolve that dispute. What it needs is precision about which limitation it is describing. The evidence above does not require that language manufacture the underlying cognitive capacity for reasoning. It requires only that language modulates which of several available strategies gets recruited, how quickly, and how automatically. On a Chomskyan picture, every human brain has the capacity for absolute spatial orientation; Guugu Yimithirr trains and rewards it, while English does not. Language was withholding the key, not building the lock.

The deeper point is that neither Whorf nor Chomsky fully accounts for the mind's capacity to exceed its own language. Cognitive neuroscience has established that high-level reasoning recruits neural circuits independent of language: Kosslyn, Ganis, and Thompson (2001) demonstrated that visual mental imagery activates the same primary visual cortex regions as actual perception, proceeding through retinotopic maps without linguistic mediation. We think spatially and analogically in ways deeper and faster than any sentence we could form. Language is not the floor of thought. It is a structuring layer — sometimes clarifying, sometimes distorting — laid over a more fluid preverbal architecture the brain never stops running. The pressing question is whether the excess that thought can achieve beyond language can be sustained, organised, and transmitted without being collapsed back into the available vocabulary. That is what this essay is working toward.

II. The Universe Refuses to Behave

Physics offers the most extreme stress-test of this idea, because physics keeps discovering things no language was built to describe. The quantum world arrived already grammatically impossible. An electron does not have a definite position until it is measured — not because we lack the instruments, but because no definite position exists to measure. English grammar, with its confident assignment of subjects, predicates, and settled locations in space, resists this at every turn.

The breakdown goes deeper than nouns and predicates. In the world's major literary and scientific languages, every sentence is anchored in tense: speakers address a present moment, reaching back into a past and forward into a future. This grammatical architecture is so fundamental that it feels less like a feature of language and more like a feature of reality. But at the boundary the physics calls the singularity, tense fails entirely. There is no before. There may be no after in any meaningful sense. The sentence "the universe began" presupposes a timeline on which the beginning is a point, but the physics suggests there is no such timeline prior to what the equations describe. We have no tense for a moment outside of time, no pronoun for an event without a subject, no verb form for a process that precedes all process.

The first case concerns the geometry of space itself. For two thousand years, Euclidean geometry was not understood as one possible description of space among many but as the only intelligible one. The axiom that parallel lines never meet was not a hypothesis — it was a conceptual necessity, so deeply embedded in spatial intuition that to deny it was not merely to make a false claim but to produce a statement that seemed to have no coherent meaning. Kant enshrined this in his critical philosophy: Euclidean space was the a priori form of outer intuition, the precondition of any possible spatial experience. When Gauss, Bolyai, and Lobachevsky independently developed geometries in which parallel lines could meet and the angles of a triangle sum to less than 180 degrees, they were generating a new conceptual vocabulary for which no shared physical referent yet existed. Gauss declined to publish his results on curved space for decades, concerned about the philosophical reception. Riemann formalised the geometry of intrinsically curved spaces of arbitrary dimension in his 1854 habilitation lecture — working, as Torretti (1978) emphasises, entirely within the space of abstract mathematical possibility, with no physical application in view. Sixty years later, Einstein needed that vocabulary to write his field equations. It was waiting for him.

A natural objection is that this episode supports the opposite of the essay's thesis: mathematics generated a new language before any physical need for it arose — precisely Wigner's (1960) "unreasonable effectiveness" at work. The objection is correct as far as it goes. But notice what the case actually shows. The delay was not in the mathematics. It was a delay in the conceptual framework physicists brought to their data. Einstein himself reported that his difficulty in completing general relativity between 1907 and 1915 lay largely in recognising, with Marcel Grossmann's help, that Riemannian geometry was the structure the physics required (Norton, 2004). Physicists trained in Newtonian mechanics simply did not have "curved spacetime" available as a thinkable description of gravity. When it became available — not through new observations but through conceptual reorientation — the data that had always been there (Mercury's perihelion precession, the bending of light) could be seen as evidence for something general relativity had predicted. The new language did not change the data. It changed what the data was evidence for.

The second case is more recent and less resolved — and faces a sharper competing viewpoint. Quantum field theory replaced the classical notion of a particle — a small, localised object with definite properties — with the concept of a quantum field, a mathematical structure defined across all of spacetime from which particle-like excitations temporarily emerge. This was not merely a technical upgrade. It required abandoning the ontological vocabulary that physics had used since Newton: the vocabulary of things, of objects, of entities with locations. A quantum field is not a thing in any ordinary sense. It is not located anywhere. It does not persist as a classical object persists. Physicists have been building and testing quantum field theories with extraordinary empirical success for nearly a century, yet the question of what a quantum field actually is, ontologically, remains genuinely open: the competing proposals include particle ontologies, field ontologies, trope ontologies, and process ontologies, none of which has secured consensus (Gasparinetti, 2025). The mathematics works. The conceptual language is still catching up.

Tracing the concept's history makes the mechanism visible. In the 1840s, Michael Faraday introduced "lines of force" to describe the influence radiating from magnets — a deliberately visual, spatial image that let him reason about action at a distance without positing an invisible medium of transmission. Faraday was not a trained mathematician; his conceptual vocabulary was diagrammatic and intuitive. In the 1860s, Maxwell translated Faraday's visual intuitions into a set of differential equations governing the electric and magnetic fields, replacing the image of lines with the abstract concept of a field as a mathematical entity distributed throughout space. The vocabulary shift mattered: Faraday's "lines" implied direction and discrete flow; Maxwell's "field" implied a continuous mathematical assignment of a value to every point in space. Physicists working in Maxwell's vocabulary could think thoughts Faraday's had not made available — in particular, that a field could carry energy independently of any visible object.

When Dirac published his relativistic wave equation for the electron in 1928, the field concept transformed again. The equation yielded solutions with negative energy that his contemporaries treated as a mathematical artefact to be discarded. Dirac instead proposed that these solutions described real physical states: a "sea" of filled negative-energy electrons, with detectable "holes" where an electron was missing. He initially identified these holes with protons before recognising, by 1931, that they must be a particle identical to the electron but positively charged — what Anderson confirmed as the positron in 1932 (Wigner, 1960). The word "particle" remained throughout this history, and it remains today across the entire Standard Model, even though the theory it describes contains no particles in any classical sense. Physicists use "particle" because it is familiar and approximately adequate for low-energy calculations. They know it misleads at a fundamental level. The conceptual vocabulary has never caught up with the mathematics, not because the mathematics is new, but because the mind still reaches for objects, locations, and things — finding instead excitations, relations, and fields for which no language built for a world of medium-sized material objects has an adequate name.

The most rigorous competing viewpoint here comes from ontic structural realism. French and Ladyman (2003) argue that QFT and quantum mechanics demonstrate not that we lack the right conceptual vocabulary for what quantum entities are, but that there simply are no objects at the fundamental level — only structures and relations, described by the mathematics. On this view, the ontological openness of QFT is not a failure of conceptual language. It is the discovery that object-vocabulary was never the right approach, and that the mathematical structure is the ontology. The proper response is not to develop better natural-language concepts of quantum fields but to abandon the demand for object-ontology entirely and accept that the equations describe all there is to describe.

This is a serious position and it would be dishonest to dismiss it. But even if French and Ladyman are right, the linguistic argument survives — because structural realism is itself a conceptual framework, expressed in natural and philosophical language, that philosophers developed, argued for, and defended against competing views over several decades. The mathematical formalism of QFT did not announce its own structural realist interpretation. Philosophers built that interpretation through sustained work that required constructing a vocabulary — "structure," "relations," "ontic" — to make the position intelligible and distinguishable from its rivals. The equations ran ahead. The concepts followed. The gap between them is exactly the gap this essay is about.

The history of physics offers an even earlier and more instructive example of the specific mechanism by which words obstruct as well as liberate. For three centuries, the word "wave" carried an implicit ontological commitment: waves are disturbances in a medium. Sound waves need air; water waves need water. When Maxwell showed in the 1860s that light is an electromagnetic wave, physicists did not ask whether waves might propagate without a medium. The word itself ruled that question out. Instead, they postulated the luminiferous ether — an invisible, all-pervading substance through which light travelled — and spent decades designing instruments to detect it.

When Michelson and Morley found no evidence of ether drift in 1887, the response was not to question the word "wave" but to save the ether: Lorentz and FitzGerald independently proposed that moving objects physically contract along their direction of motion, precisely enough to explain the null result. The linguistic commitment to a medium held the entire field inside a framework the physics had already outgrown. Einstein's 1905 paper on special relativity did not refute the ether. It dissolved it — by replacing the conceptual vocabulary so that "electromagnetic wave propagating without a medium" became not just permitted but meaningless to deny (Cassini & Levinas, 2019). The problem had been the word, and the solution was a new language.

Cosmology presents an even more vertiginous confrontation. Take the Big Bang — two of the most loaded words in science. The name conjures an explosion, a before and after, a moment when everything began. But this framing is almost certainly misleading. A 2025 survey of physicists gathered for the "Black Holes Inside and Out" conference in Copenhagen found that the only statement about the Big Bang to gain majority approval — from 68% of respondents — was simply that the universe evolved from a hot, dense early state. The stronger claim that it represented an absolute beginning of time did not reach consensus (Chen, Halper, & Afshordi, 2025). The phrase "Big Bang" has a before built into it. The physics may not.

A paper published in Physical Review Letters in 2026 shows how far the mathematics can run beyond ordinary language. Liu, Quintin, and Afshordi demonstrated that within a quadratic gravity framework — one whose coupling constants remain well-behaved at arbitrarily high energies — the physics can be described continuously through the region that classical general relativity treats as a singularity (Liu, Quintin, & Afshordi, 2026). There is no point at which the equations break down and demand a "beginning." The formalism does not require one. The word does.

What does it mean to call something a beginning if nothing begins there? The word carries the full weight of a narrative grammar — stories have openings, causes precede effects, time has a first entry. When physics describes a region that ordinary language insists on calling the start, the difficulty is not only technical. It is linguistic. We do not yet have a word for "the mathematical region our narrative grammar calls a beginning but which the physics treats as continuous." The equations reach past the available vocabulary. The vocabulary keeps reaching back.

III. When Experts Cannot Agree on a Name

The Copenhagen survey of 2025 makes this concrete. Some of the world's most accomplished physicists, gathered around the questions that define the frontier of the discipline, could not reach majority agreement on frameworks that dominate textbooks and grant applications. The cosmological constant, cosmic inflation, string theory — while each commanded support from a significant minority, none won a majority (Chen et al., 2025). This should give pause before describing any of these frameworks as consensus science.

It is physics being honest about the edge of its own language. The words we have — "singularity," "inflation," "the beginning" — are placeholders: they name a gap shaped like a thing we have not yet understood. The danger comes when we mistake the name for the understanding, when the placeholder calcifies into a concept and we stop seeing the gap it was always pointing at.

The name "dark matter" is perhaps the most consequential current example. The observations it describes — anomalous galactic rotation curves, gravitational lensing, large-scale structure — do not require additional matter; they require additional gravity. Modified gravity theories such as MOND make the same predictions in many regimes. Yet "dark matter" as a name commits the field to a substance, and the result has been decades of particle searches rather than a more equal exploration of gravitational alternatives (Hossenfelder & McGaugh, 2018). The name preceded the understanding. The name may now be constraining it.

The history of black hole physics offers a complementary example of how a name can both liberate and constrain. When Karl Schwarzschild published his solution to Einstein's field equations in 1916, the objects it described were called "frozen stars" for decades — a name encoding the experience of a distant outside observer, to whom infalling matter appears to slow and freeze at the event horizon. The name was accurate in one limited sense. But it committed physicists to thinking of these objects as a kind of extreme stellar endpoint — material things in suspension — rather than as geometric features of spacetime. Einstein himself wrote a 1939 paper arguing that physical processes would prevent such objects from ever forming. Most leading relativists shared this resistance. When John Wheeler popularised the name "black hole" around 1967, replacing "completely collapsed gravitational object" and "frozen star" alike, the shift was not merely terminological (Herdeiro & Lemos, 2019). A black hole is not a star that stopped. It is a region — a feature of spacetime geometry, not a material object at all. The new name enabled an entirely new research programme: one focused on event horizons, information loss, Hawking radiation, and the topology of spacetime, questions that the "frozen star" framing had made nearly unaskable.

The Chandrasekhar limit shows the same mechanism at work more personally. In 1931, the young physicist Subrahmanyan Chandrasekhar calculated, using special relativity and quantum mechanics, that a non-rotating star above approximately 1.4 solar masses cannot be supported against gravitational collapse — it must continue collapsing beyond the white dwarf stage. Eddington, then the most authoritative astrophysicist in the English-speaking world, rejected the result publicly at the Royal Astronomical Society in 1935, declaring that "there should be a law of nature to prevent a star from behaving in this absurd way." His vocabulary of stellar structure, built around stable equilibria and well-behaved material states, had no room for a star that collapsed without limit. Chandrasekhar's mathematics was correct. It was conceptually inadmissible within Eddington's framework. The broader community largely sided with Eddington, and Chandrasekhar's insight was marginalised for nearly three decades, until the revival of relativistic astrophysics in the 1960s (Herdeiro & Lemos, 2019). The entrenched vocabulary was not neutral. It encoded prior commitments about what kinds of things could physically exist — and those commitments delayed recognition of a correct result for a generation.

This is the precise opposite of the Kuuk Thaayorre's remarkable precision. Their language encoded a real directional structure of the world with such fidelity that it built a navigational faculty into daily cognition — the vocabulary matched the territory. At the cosmological frontier, no vocabulary yet matches the territory, and the ones we have may actively mislead, drawing our models back toward human-scale intuitions about before and after, inside and outside, beginning and end.

IV. Building Better Words

There is a productive response to all of this, and it is simpler than radical linguistic reform. Think of a builder who knows that scaffolding is not the building — who erects it deliberately, uses it practically, and removes it at the right moment without grief. "Singularity" and "beginning" are scaffolding. They hold the inquiry upright while better conceptual structures are built. The error is not using them. The error is forgetting they are temporary.

Albert Einstein spent nearly a decade searching before completing his general theory of relativity in November 1915, a theory that overturned Newtonian notions of absolute space and time and replaced them with a geometry of curved spacetime. The search began not with equations but with a spatial image. At around sixteen, Einstein imagined riding alongside a beam of light and noticed that classical physics produced an incoherent result: Maxwell's equations required light to travel at a constant speed regardless of the observer; Newtonian mechanics said a rider moving at the same speed should see a frozen wave. No existing concept resolved this. Einstein held the image for nearly a decade until he found the conceptual reorientation that could (Norton, 2004). Imagination had outrun language: it suspended the demand for a grammatical resolution, held the contradiction as a spatial fact, and kept it available for thinking. Language would have required a premature answer. The image was freer.

Mathematics does not merely escape language's constraints — in certain conditions it actively outruns them, reaching true conclusions that no sentence could have reached first. Three cases look, at first glance, like counterexamples to everything argued so far and deserve confronting directly.

Isaac Newton invented the calculus not because existing language failed him, but because no symbolic system existed for reasoning about continuously changing quantities at all. There was no word, no sentence, no grammar — not even a faulty one — for instantaneous rates of change; the entire conceptual territory was blank. Newton did not work around a linguistic ceiling. He built the first floor.

Paul Dirac's case is sharper still. In 1928 he wrote down a relativistic wave equation for the electron and found, embedded in its own algebra, solutions corresponding to negative energy — a result his contemporaries treated as a mathematical embarrassment to be discarded. Dirac instead trusted the equation and proposed the concept of a "hole" in a filled sea of negative-energy electrons: a detectable absence that would behave as a positively charged particle. Carl Anderson detected exactly such a particle — the positron — in 1932, confirming a physical entity that the mathematics had implied a full three years before any physicist had a word for it (Wigner, 1960).

Einstein's general relativity belongs in the same company: the field equations predicted gravitational lensing, the bending of starlight around mass, years before the 1919 eclipse expedition that confirmed it — a phenomenon for which ordinary language had no pre-existing concept at all.

What these three cases share, and what distinguishes them from the singularity, the ether, or dark matter, is the direction of travel. In each, the mathematics arrived first and the concept — and eventually the word — followed behind it, because no entrenched word stood in the way. There was no established noun for "negative-energy hole" competing with Dirac's algebra; no established description of continuous change competing with Newton's fluxions.

Natural language becomes an obstacle specifically when a word already exists, already feels adequate, and already carries an implicit conceptual commitment that the new physics must work around or quietly violate. "Wave" already meant medium-borne disturbance when Maxwell needed it not to. "Beginning" already meant a point with a before when the singularity required there to be none. "Dark matter" already commits to a substance the observations may not require. Newton, Dirac, and Einstein's lensing prediction were not obstructed by entrenched vocabulary because none existed yet to obstruct them. The danger is not the absence of words. It is the presence of the wrong ones, arriving early and overstaying their welcome.

Mathematical language has real advantages here. Equations make no commitment to a narrative direction. The formalism of quadratic gravity contains no word for "beginning" and is therefore not seduced by one — which is precisely why Liu, Quintin, and Afshordi could write a theory that runs cleanly through the region ordinary language insists on calling a singularity. The mathematics was not infected by the metaphor.

Natural language can approach this discipline too, but it requires conscious effort: holding the ordinary description and the corrective one simultaneously, without collapsing them into each other. This is physical intuition — the trained ability to navigate the gap between what the equations say and what any sentence will make them sound like. A community that builds more precise scientific metaphors — that stops treating the Big Bang as an explosion in pre-existing space, or quantum states as hidden classical realities — thinks more clearly about these problems, not merely communicates more carefully. The words matter because the words are part of the thinking.

V. The Limits of the Framework

If words shape the thoughts we can reach for, then the frameworks within which science operates are not merely passive vehicles for communicating knowledge — they are active structures that shape which hypotheses seem natural, which questions seem worth asking, and which answers count as explanations. This influence is sharpest at the frontier, where what Sellars called the "manifest image" of ordinary experience provides no reliable guide (Sellars, 1962; Vacura, 2015).

Ludwig Wittgenstein was among the first to press this philosophically. In the Tractatus Logico-Philosophicus he argued that the limits of one's language define the limits of one's world — that the structure of what can be said constrains what can be clearly thought (Wittgenstein, 1922). His later Philosophical Investigations refined rather than abandoned this: meaning arises through use within particular social practices — "language-games" — and new practices can open new conceptual territory (Pinto, 2024). Crucially, this claim does not require that language manufacture the underlying cognitive capacity for thought; it requires only that the language-game in use determines which inferential moves feel available. Language constrains thought, but new language can expand it.

The cognitive linguist George Lakoff, working with Mark Johnson, extended this insight into the structure of abstract thought itself. Their central argument is that abstract concepts — including scientific ones — are not free-floating logical structures but are organised through embodied conceptual metaphors: systematic mappings from familiar physical domains onto abstract ones (Lakoff & Johnson, 1980). Consider how naturally we speak of time running out, of ideas being in the air, of understanding being grasped. We do not consciously choose these metaphors. They arrive with the concept, pre-attached, carrying all their physical implications with them. Time runs out — so it is a finite resource, something that can be saved or wasted. A theory is built — so it can be undermined from below, or have its foundations shaken. These are not rhetorical decorations. They are load-bearing structures in the conceptual architecture, determining what follows and what does not. When physicists speak of the universe "expanding," of particles "tunnelling" through barriers, or of space "bending" around mass, they are deploying exactly this kind of embodied metaphor — making some theoretical moves feel intuitive and others feel impossible before a single equation is written. Lakoff and Núñez (2018) show that even mathematics is not exempt: infinity is conceptualised as a completed journey, limits as an approach toward a destination, sets as containers. The scaffolding goes all the way down.

Wilfrid Sellars gave this a specifically scientific dimension. His distinction between the manifest image and the "scientific image" produced by theory identifies a permanent source of friction (Sellars, 1962). The two frameworks are not simply different descriptions of the same facts; they are organised differently and come into genuine conflict. Consider the table you are sitting at. The manifest image says it is solid, brown, smooth, stationary. The scientific image says it is almost entirely empty space — a lattice of nuclei separated by vast distances relative to their size, surrounded by probability clouds of electrons, the whole structure vibrating at frequencies far beyond perception, the "solidity" an emergent consequence of electromagnetic repulsion between electron clouds. Neither description is false. They are incommensurable: built from different categories, answering different questions, using different criteria for what counts as a thing at all. When physics ventures beyond the manifest image into quantum superposition, spacetime curvature, or a cosmology without a temporal beginning, it is not adding new facts to a neutral vocabulary. It is entering territory the manifest image was never built to map (Vacura, 2015).

Thomas Kuhn made the historical dimension explicit. In The Structure of Scientific Revolutions he argued that major scientific advances transform the conceptual vocabulary through which observations are made and interpreted (Kuhn, 1962). When a scientific lexicon is replaced in a paradigm shift, the old vocabulary cannot express the new ideas without fundamental distortion. Here is a concrete instance: before 1543, astronomical observations were recorded within a framework in which the Earth did not move. Full stop. It was not that movement was considered and rejected; the concept of a moving Earth was not a live option the Ptolemaic system could evaluate. The categories were wrong before the measurements were taken. Copernicus did not offer a better answer to an existing question. He changed the question. The current difficulty in agreeing on what the Big Bang was — an absolute origin, a boundary condition, or a region the mathematics simply passes through — may signal that cosmology is approaching exactly this kind of threshold. We are still asking the old question. We may need to change it.

What distinguishes natural language, mathematics, and specialised symbolic systems is the kind of constraint each imposes. Natural language is the most pervasive and least visible: it operates before scientific thinking begins, shaping which questions feel natural, which analogies feel apt, and which answers seem satisfying — Whorf's dissection of nature along linguistic lines (Whorf, 1956) at its most consequential. Mathematical language escapes narrative grammar but substitutes structural assumptions of its own — continuity, smoothness, the algebra of real numbers, Euclidean geometry — frameworks that make certain theories feel natural and others feel contrived. Specialised symbolic systems such as Feynman diagrams occupy a third register: genuinely representational, as Meynell (2008) argues, shaping which physical processes physicists visualise as possible and which they reach for first in calculation. Each register carries its own probabilistic nudge, its own tilt toward certain configurations of the thinkable.

A common objection is that mathematics transcends these ambiguities — and it deserves its strongest possible statement before being answered. The objection has two layers, and the second is harder than the first.

The first layer: mathematics is simply more precise than natural language. This is correct. Emmy Noether's 1918 theorem established that every continuous symmetry of a physical system corresponds to a conserved quantity: time-translation symmetry gives conservation of energy; spatial-translation symmetry gives conservation of momentum. This is not a fact that could have been found by thinking in words. It required a formal language in which symmetry and conservation could be related algebraically, stripped of narrative content. Quantum field theory describes particles as excitations of fields in ways that have no ordinary-language equivalent. The mathematics went somewhere natural language cannot follow, and found something real.

The second layer is stronger. Mathematics does not merely describe existing reality more precisely — it generates new ontologies that pre-exist our conceptual vocabulary for them. Riemann's geometry of curved spaces sat waiting for sixty years before Einstein needed it. Dirac's negative-energy solutions described the positron before anyone had a word for it. On this view, mathematics is not a language at all in the Whorfian sense. It is a conceptual technology that operates independently of the human mind's linguistic architecture. The sharpest form of this objection is mathematical Platonism: the view that mathematical structures exist mind-independently, and that mathematicians discover rather than invent them. If Platonism is true, then the mathematical scaffolding of Lakoff and Núñez is irrelevant — the structures pre-exist any human cognitive representation of them, and our metaphor-laden ways of thinking about infinity or continuity are merely the access routes to a domain that would exist even if we used different metaphors. Natural language imposes a ceiling, but mathematics is already beyond it by definition.

This is the strongest version of the objection. It should not be dismissed. But it does not make mathematics bias-free — it relocates the problem from ontology to epistemology. Even granting the Platonist premise that mathematical structures exist independently of the mind, the question of which structures physicists reach for, which they find natural, which they find elegant, and which they apply to which physical situations remains a thoroughly human affair — shaped by training, community norms, and precisely the probabilistic nudges this essay has been documenting. Riemann's geometry sat unapplied to physics for sixty years not because it was unknown but because no one had built the conceptual bridge to gravity. That bridge was cognitive and linguistic as much as mathematical. Lakoff and Núñez (2018) need only the defensible claim that the human mind's access to and application of mathematical structures is mediated by conceptual metaphors — not the stronger claim that mathematics is invented. Symbols do not interpret themselves. The equations of quadratic gravity pass through what we call a singularity; the word "singularity" is still needed to identify which physical problem the equations address. Mathematics reduces linguistic ambiguity. It does not complete conceptual interpretation.

Mathematics itself carries probabilistic nudges. When British mathematicians refused to adopt Leibniz's more expressive calculus notation out of loyalty to Newton, their mathematics fell measurably behind the European continent for the better part of a century (Li, 2024). Notation is not neutral — the same mechanism Boroditsky documented in natural language, operating inside the formalism. The choice of mathematical structures runs deeper: most tools physicists reach for assume spacetime is continuous and infinitely divisible, a mathematical habit rather than an established physical fact. The quadratic gravity framework works precisely because it adopts a different structure, one that remains UV-complete where the standard toolkit breaks down.

Mathematical communities also develop socially enforced aesthetic norms, and these guide research as reliably as any grammatical habit (Jevtić, Kostić, & Maksimović, 2024). String theory's hold on theoretical physics over several decades was sustained in significant part by mathematical elegance in the near-absence of experimental confirmation — a pattern Ritson and Camilleri (2015) documented as a sociological as much as a scientific phenomenon.

The task of the scientist is therefore not to escape these systems — that is impossible — but to hold them consciously enough to notice when they are steering rather than serving. Mathematics is a less linguistically biased language, not a bias-free one. Equations are not bound to a subject-predicate structure, do not force a before and after, and can describe relationships symmetric in time in ways that sentences cannot. So mathematics genuinely escapes some of the traps of ordinary language. But it brings its own image along. Scientific progress depends not only on the accumulation of data but on the ongoing development of conceptual vocabularies capable of organising what is found — for advances in knowledge often begin, as Kuhn recognised, with advances in the language through which knowledge becomes thinkable.

VI. A Theory of Conceptual Lag

The evidence assembled in this essay — from Boroditsky's priming effects to Bohr's insistence on classical description, from the ether's long afterlife to the unresolved ontology of quantum fields — converges on a pattern. This section proposes a framework for naming and organising it: not a finished theory, but a structured hypothesis that generates predictions about where scientific communities get stuck and how they break free.

The framework has three components: the mechanism of Conceptual Lag, the theory of Three Registers, and the threshold concept of Escape Velocity.

Conceptual Lag is the systematic delay between a mathematical or physical discovery and the development of a conceptual vocabulary adequate to interpret it — distinguished from ignorance and from the lag of experimental confirmation. A community in Conceptual Lag may fully understand the mathematics and have accumulated abundant data. The lag is structural: when new results arrive, they are inevitably first interpreted through the entrenched vocabulary, which shapes what questions are asked, what counts as a satisfying answer, and what appears anomalous. The new result is absorbed as an outlier within the existing framework rather than treated as a demand for a new one. The ether-saving after Michelson-Morley is Conceptual Lag in its purest form — the mathematics had already moved; the concepts had not. Bohr's insistence on classical language is a subtler instance. The measurement problem in quantum mechanics — still unresolved after a century of correct predictions — is arguably Conceptual Lag in its most extreme form: the mathematics works, but no conceptual vocabulary yet makes its predictions intelligible without generating apparent paradoxes.

Conceptual Lag makes three predictions. First: an entrenched vocabulary will be used to interpret new discoveries for as long as possible, generating increasingly elaborate theoretical structures to preserve it — as Ptolemaic astronomers added epicycles rather than questioning circular motion. Second: the entrenched vocabulary will generate false problems — problems arising not from the physics but from applying the wrong conceptual system. The singularity as an "absolute beginning" is a false problem of this kind: tense-grammar applied to a regime where tense has no referent. "Where was the universe before the Big Bang?" is not a cosmological question. It is a grammatical error. Third: resolution requires not new data but conceptual reorientation — a switch in which vocabulary is taken as primary. This is why Kuhn was right that paradigm shifts are revolutions rather than accumulations; where interpreters dispute his stronger claim of incommensurability, the Conceptual Lag framework agrees with the weaker position: the new vocabulary reinterprets the old data, which is not lost but relocated.

The Three Registers is the second component of the framework. The essay has distinguished three levels of representational system — natural language, mathematics, and specialised symbolic systems such as Feynman diagrams — and shown that each constrains thought differently. The theory proposes that each register also has a characteristic lag time and a characteristic failure mode. Natural language lags longest, typically on the order of decades to centuries (the ether lasted three centuries; the Earth-centred cosmos, longer), and its characteristic failure mode is ontological commitment: it binds physicists to a vocabulary of things, substances, locations, and causal agents that the physics may have already abandoned. Mathematical language lags less — on the order of years to decades — but its characteristic failure mode is interpretive gap: the formalism generates correct predictions while leaving the question of what it means, ontologically and conceptually, genuinely open. Symbolic systems lag least, but their characteristic failure mode is scope limitation: they are designed for specific problem classes and break down when generalised beyond them, as Feynman diagrams break down in strongly coupled regimes where perturbation theory fails.

The theory predicts that scientific revolutions typically require register-switching: moving the primary site of inquiry from one register to another. Newton's invention of the calculus was a switch from natural-language description of motion to a mathematical register for which no natural-language equivalent existed. Einstein's shift to Riemannian geometry was a switch within the mathematical register — from one mathematical vocabulary (Euclidean differential geometry) to a more general one (Riemannian manifolds). Bohm's proposed Rheomode would, if completed, represent a switch within the natural-language register itself: replacing a noun-based grammar with a verb-based one to eliminate the ontological commitment to static objects. The theory further predicts that partial register-switches — in which a community adopts the new mathematics while retaining the old natural-language vocabulary — are unstable: they generate the phenomenon of Conceptual Lag, in which the entrenched vocabulary constrains interpretation even when the formalism has moved beyond it. This is the current situation of quantum mechanics, where the mathematical register (quantum field theory, the path integral formulation, decoherence theory) has far outrun the natural-language register (particles, waves, collapse, observation), generating a century of interpretive stalemate.

Escape Velocity is the third component. The term is borrowed deliberately from orbital mechanics: just as a body must reach a critical threshold of kinetic energy to escape a gravitational field entirely rather than falling back, a scientific community must reach a critical threshold of conceptual density — a sufficient number of practitioners sharing a sufficient number of terms and inferences in the new vocabulary — before a new register becomes self-sustaining rather than being continually reabsorbed into the old. Below Escape Velocity, individual insights are translated back into the dominant vocabulary and domesticated: Everett's relative-state formulation was ignored for two decades not because the physics was wrong but because the community had not achieved Escape Velocity in the non-classical conceptual register required to evaluate it on its own terms. Rosenfeld's characterisation of it as "hopelessly wrong" (Osnaghi, Freitas, & Freire, 2009) is precisely what Escape Velocity failure looks like: a verdict rendered in one vocabulary on work that requires another. Above Escape Velocity, the new register becomes self-reinforcing: practitioners train successors in the new vocabulary, the new vocabulary generates its own problems and solutions, and the old vocabulary becomes available for what it always was — a useful approximation in familiar regimes.

Escape Velocity is not achieved by individual insight alone, which is why Einstein's decade of searching required Grossmann's mathematics, why Everett's framework required DeWitt's reformulation and Wheeler's advocacy before it gained traction, and why the Copernican revolution required Galileo, Kepler, and Newton before it was irreversible. The theory predicts that the rate of conceptual change is not primarily a function of the quality of individual ideas but of the social conditions under which new vocabularies can be learned, practised, and transmitted — which means that the philosophy of science, education in physics, and the sociology of scientific communities are not peripheral to scientific progress but constitutive of it.

The framework is consistent with the existing literature rather than in conflict with it. Kuhn's paradigm shifts are instances of Escape Velocity being achieved after a period of Conceptual Lag. Sellars's manifest/scientific image divide describes the characteristic failure mode of the natural-language register when confronted with scientific image concepts that exceed it. Wittgenstein's language-games are local vocabularies that define what inferences feel available within a register — switching registers is switching language-games at the level of the representational system itself. Boroditsky's experimental findings provide the cognitive-science substrate: Conceptual Lag is grounded in the probabilistic nudges that entrenched vocabularies exert on the habitual patterns through which scientists perceive, interpret, and communicate their data.

The most productive intervention in a stalled scientific field is not more data — data interpreted through the entrenched vocabulary will be absorbed by it — but the deliberate cultivation of a new vocabulary in a new register, taught to enough practitioners to exceed the Escape Velocity threshold. The frontier questions of contemporary physics — the interpretation of quantum mechanics, the ontology of quantum fields, the conceptual status of the Big Bang singularity, the nature of dark matter — are not primarily empirical problems awaiting more powerful instruments. They are, in the precise sense this framework describes, problems of Conceptual Lag: the instruments have already outrun the words.

VII. The Sky Remains

The universe is larger than any of the languages we have developed to describe it. So is the problem. The Conceptual Lag framework identifies where communities get stuck, but it does not guarantee escape — and at the furthest frontier, there may be a kind of understanding that exceeds not just natural language but any transmissible representational system.

The Kuuk Thaayorre built a compass into their language and carried it in their heads. The physicists at Copenhagen could not agree on what words to use for what they had found. Both facts illuminate the same relationship: between the structures we can speak and the structures that are there. Sometimes language is a precision instrument, carving cognition to match the world with extraordinary fidelity. Sometimes it is a blunt inheritance — a tool built for an older world that keeps snagging on the new one.

There is a darker reading of the evidence assembled here, and it deserves to be stated plainly. If language shapes thought as persistently as the research suggests — if the grammar of time nudges cognition, if mathematical notation steers a field for a century, if aesthetic norms sustain entire research programmes without experimental support — then language begins to look less like a clumsy tool we constantly outgrow and more like a rigid ceiling we press against without knowing it is there. On this reading, the history of science is not only a story of imagination overcoming ignorance. It is also a story of frameworks quietly setting the terms of what imagination was allowed to reach for, and how much it never reached at all.

The history of quantum mechanics offers the sharpest illustration. Niels Bohr insisted that quantum phenomena could only be described in classical terms — not as a temporary measure, but as a permanent constraint. Any unambiguous account of a measurement, he argued, must be framed in the language of Newton and Maxwell, indefinitely. Several commentators have since argued that this insistence foreclosed non-classical conceptual development for decades, trapping the interpretation of quantum mechanics within a vocabulary the theory itself had already outgrown (Zinkernagel, 2016). The ceiling was not imposed from outside. It was built into the language in which the questions were being asked.

The consequences were concrete. Hugh Everett III, a graduate student at Princeton, developed his relative-state formulation of quantum mechanics in 1957 — proposing that the universal wave function never collapses and that quantum superpositions branch into all possible outcomes simultaneously. The formulation required no classical observer and no collapse: it operated entirely within quantum mechanical language.

Bohr's response, following Everett's visit to Copenhagen in 1959, was rejection: he regarded the approach as violating his prohibition on treating physical reality as fundamentally quantum mechanical rather than classical, and his collaborator Léon Rosenfeld characterised Everett's ideas as "hopelessly wrong" (Osnaghi, Freitas, & Freire, 2009). The framework was largely ignored for nearly two decades.

The mathematics was correct. What Everett had built was a new language — and the existing community could not initially hear what it said. The ceiling was not resolved by better data. It was resolved by a conceptual vocabulary that did not need the old constraints to function.

Kuhn's paradigm analysis makes the structural point general. Before Copernicus, the question "why does the Earth move?" could not be asked in a way that made astronomical sense — the Ptolemaic lexicon treated terrestrial rest as a starting assumption, not a hypothesis. Before Einstein, the question "what would happen if there were no fixed reference frame?" had no foothold in Newtonian mechanics. In each case, the ceiling was not a prohibition but an absence: a region of thought the framework simply did not make available (Kuhn, 1962).

Yet ceilings do get broken. Einstein held his light-beam image for nearly a decade until the mathematics arrived to express what it implied. Bohr's classical constraint was eventually contested, and alternative interpretations have since developed precisely the non-classical vocabularies Bohr declared impossible: relational quantum mechanics, in which quantum states are defined relative to an observer rather than absolutely (Rovelli, 1996), and QBism, which treats the quantum state as an agent's personal probability assignment rather than a description of observer-independent reality (Fuchs, Mermin, & Schack, 2014).

Leibniz's notation displaced Newton's. Liu, Quintin, and Afshordi wrote a theory that passes cleanly through a region ordinary language insists on calling an end. In each case, the framework that felt like a ceiling turned out to be a floor — a foundation for something larger, once someone found the door.

The word "beginning" cannot hold the sky. Neither, perhaps, can "singularity" or "inflation" or the many other nouns pressed into service at the frontier. And this may be the deepest diagnosis: that it is not merely which words we choose but the grammatical architecture they inhabit that limits us. The subject-verb-object structure shared by most Indo-European languages encodes a world of separate, static objects acting on one another — things, not flows. It is precisely the wrong grammar for quantum mechanics, where there are no isolated objects, only entangled relations. It is precisely the wrong grammar for cosmology at the singularity, where there is no subject that acts and no object that receives.

The physicist David Bohm recognised this in 1980, proposing what he called the Rheomode: an experimental language built entirely on verbs and flows rather than nouns and static objects (Bohm, 1980). Where ordinary language asks "It is raining" — implying an entity that does the raining — the Rheomode would enact raining as pure process, without a subject. Bohm did not complete it. It remains a provocation rather than a working language. But the provocation points in the right direction: to understand the universe at its most fundamental level, we may not need better words. We may need a different kind of sentence.

Yet perhaps the honest path forward is not a better grammar but the courage to stop translating altogether. The Schrödinger equation predicts measurement outcomes with extraordinary precision. What it does not settle is the ontological interpretation — what is physically happening between measurements. A century of debate has not converged on an answer, not because physicists lack data, but because every proposed interpretation must be rendered into a noun-verb language the mind can navigate, and the mathematics resists each rendering without remainder. The sky may already be described in the formalism. What lags behind is the mind's insistence that understanding must eventually arrive as a sentence. What if it does not?

Michael Polanyi (1966) argued that all human knowledge has a tacit dimension: we always know more than we can tell. The skilled surgeon knows something a textbook cannot fully capture. The experienced physicist has intuitions about which calculations to trust that exceed any explicit rule. Applied to the frontier of physics, the implication sharpens. If the deepest understanding of quantum fields, of pre-singularity cosmology, of the quantum-to-classical transition is tacit in Polanyi's sense — if it consists in trained geometric intuition, in the ability to navigate the space of solutions without a linguistic map — then progress at the frontier is not primarily a matter of better papers or clearer explanations. It is a matter of apprenticeship: trained practice transmitted through demonstration, proximity, and repetition, the way a master passes a craft to a student, not through a manual but through presence. The knowledge is real. It cannot be fully encoded. And that makes it fragile — dependent on an unbroken chain of practitioners who can train the next generation in the embodied intuition the formalism requires but cannot itself supply.

This has a further implication that is harder still. If tacit understanding at the physics frontier cannot be fully formalised, it cannot be fully verified. We can confirm that a physicist's predictions are correct without being able to confirm that what they understand is what they think they understand — or that two physicists who agree on every prediction share the same understanding of what those predictions mean. The measurement problem in quantum mechanics has been unresolved for a century not because physicists disagree about the predictions, which are not in dispute, but because they cannot agree on what the formalism means — and meaning is precisely what tacit knowledge cannot supply to those outside the practice.

Wittgenstein's later work completes the picture. In the Philosophical Investigations he argued that meaning requires shared criteria — that the possibility of following a rule depends on a community of practice in which there is a shared basis for saying a rule has been followed correctly or not (Wittgenstein, 1953). A practice without shared criteria is not a language at all. The same applies to understanding: scientific understanding that cannot be transmitted, even partially, even through the elaborate indirection of mathematics, diagrams, and trained intuition, loses its claim to be scientific. Science depends on a community able to scrutinise, contest, and build upon what any member claims to know. Tacit knowledge can be transmitted — through apprenticeship, demonstration, and the long formation of working physicists — but only to those who undergo the training, not broadcast to all. At the extreme frontier, where even the mathematical intuitions required to navigate the formalism demand years of formation, the community able to genuinely evaluate a claim may shrink to a handful of people in the world.

If understanding does not arrive as a sentence, it does not disappear. It persists, embedded in skilled practice and trained intuition. But it becomes invisible to outside scrutiny, fragile in transmission, and inaccessible to all but the initiated. The frontier of physics may already be crossing this threshold — not because the physics is wrong, and not because physicists are obscurantist, but because the conceptual vocabulary required to interpret the mathematics has not yet been built, and the gap between the formalism and available language is wide enough that only those who have internalised the formalism directly can navigate it.

The real question is not whether we can speak the sky. It is whether an understanding that cannot be spoken can still be shared — and whether a science that cannot share its deepest understanding remains, in any meaningful sense, science at all.

The answer, this essay proposes, is that sharing tacit understanding is not impossible — only different. It is closer to what happens in an atelier: the master physicist works through a problem aloud while the student watches, absorbing not an argument but a way of moving through conceptual space. It is what Keats called negative capability — the capacity to remain in uncertainty and doubt without irritable reaching after fact and reason. That capacity, to hold a paradox without demanding it become a sentence, is not mysticism. It is the cognitive move Einstein made when he held the light-beam image for a decade. It is what every physicist does who has learned to trust an equation over the available words for it.

The concrete implication for scientific education follows directly. If the deepest understanding at the frontier is tacit and imagistic rather than propositional, training that produces physicists capable of working there cannot consist entirely of lectures, textbooks, and problem sets — tools optimised for explicit, propositional knowledge. It must include the cultivation of spatial imagination: training students to visualise four-dimensional manifolds, to feel the topology of a Hilbert space, to hold a superposition as a simultaneous spatial reality rather than a notational shorthand. It must include the history of conceptual revolutions — not as a catalogue of discoveries but as a study in how communities got trapped in their own language and how they broke free. And it must cultivate what Bohm (1980) identified as the verb-thinking beneath the noun-world: attention directed toward process, flow, and relation rather than objects, properties, and states. These are not soft supplements to a rigorous curriculum. Given what this essay has argued, they may be the curriculum's most rigorous requirement.

The emergence of large language models adds an unexpected dimension to this problem. Systems trained on vast corpora of human language inherit, at scale, every probabilistic nudge that natural language exerts on thought. An AI trained on physics papers learns to generate fluent descriptions of quantum fields, singularities, and dark matter — descriptions shaped by exactly the conceptual vocabularies this essay has been examining. It learns that "particles" are things with positions, that the Big Bang was an "event," that dark matter is a "substance." It inherits the Conceptual Lag of every physicist who ever wrote about these topics, compressed and amplified across billions of tokens. What it does not inherit — and cannot, from linguistic training alone — is the tacit, imagistic, preverbal understanding that this essay argues is essential to working at the frontier. A language model can produce Bohr's interpretation of quantum mechanics and Everett's with equal fluency. It cannot notice which one is conceptually trapped. The gap between linguistic competence and conceptual understanding is, in artificial intelligence, not a technical limitation awaiting a better architecture. It is a demonstration of the essay's central argument: that the most important thinking happens in a register that language can report on but cannot fully capture.

What kind of minds are we? What kinds of representation can we build? Whether the gap between the structure of reality and the structure of human knowledge is a temporary feature of an immature discipline, or a permanent feature of the cognitive situation of beings who evolved to name objects in a world of medium-sized things — that remains genuinely open. But the frontier of physics is not waiting for the answer. It has already moved into the region where the instruments outrun the words, and the question of how understanding can be sustained and transmitted without language is no longer philosophical. It is practical. It is urgent.

In 1990, at the request of Carl Sagan, the Voyager 1 spacecraft turned its camera back toward the inner solar system from a distance of approximately six billion kilometres and took a photograph. Earth appears in it as a fraction of a single pixel — a pale blue dot suspended in a beam of scattered sunlight. Every word ever spoken, every language ever built, every conceptual framework within which any human being has ever thought — all of it sits on that speck. Language was shaped by that speck. By the pressures of that speck. By the medium-sized objects, the social negotiations, the sensory experiences available on that speck. The universe is not a speck. It is the darkness around it, stretching in every direction without a name. The sky cannot be held in a word. That is not a failure of science. It is science's most important discovery about itself — and the beginning, not the end, of the work.

References

  1. Boroditsky, L. (2001). Does language shape thought? Mandarin and English speakers' conceptions of time. Cognitive Psychology, 43(1), 1–22.
    https://doi.org/10.1006/cogp.2001.0748 · ScienceDirect
  2. Boroditsky, L., & Gaby, A. (2010). Remembrances of times east: Absolute spatial representations of time in an Australian Aboriginal community. Psychological Science, 21(11), 1635–1639.
    https://doi.org/10.1177/0956797610386621 · SAGE Journals
  3. Levinson, S. C. (1997). Language and cognition: The cognitive consequences of spatial description in Guugu Yimithirr. Journal of Linguistic Anthropology, 7(1), 98–131.
    https://doi.org/10.1525/jlin.1997.7.1.98 · Wiley
  4. Sapir, E. (1929). The status of linguistics as a science. Language, 5(4), 207–214.
    https://doi.org/10.2307/409588 · JSTOR
  5. Whorf, B. L. (1956). Language, thought, and reality: Selected writings of Benjamin Lee Whorf (J. B. Carroll, Ed.). MIT Press.
    https://archive.org/details/languagethoughtr00whor
  6. Chomsky, N. (1959). A review of B. F. Skinner's Verbal Behavior. Language, 35(1), 26–58.
    https://doi.org/10.2307/411334 · PhilPapers
  7. Chomsky, N. (1965). Aspects of the theory of syntax. MIT Press.
    MIT Press
  8. Pinker, S. (1994). The language instinct: How the mind creates language. William Morrow.
    stevenpinker.com
  9. Deutscher, G. (2010). Through the language glass: Why the world looks different in other languages. Metropolitan Books.
    Internet Archive
  10. Regier, T., & Kay, P. (2009). Language, thought, and color: Whorf was half right. Trends in Cognitive Sciences, 13(10), 439–446.
    https://doi.org/10.1016/j.tics.2009.07.001 · PubMed
  11. Meynell, L. (2008). Why Feynman diagrams represent. International Studies in the Philosophy of Science, 22(1), 39–59.
    https://doi.org/10.1080/02698590802280902 · Taylor & Francis
  12. Osnaghi, S., Freitas, F., & Freire, O. (2009). The origin of the Everettian heresy. Studies in History and Philosophy of Modern Physics, 40(2), 97–123.
    https://doi.org/10.1016/j.shpsb.2008.10.002 · ScienceDirect
  13. Winawer, J., Witthoft, N., Frank, M. C., Wu, L., Wade, A. R., & Boroditsky, L. (2007). Russian blues reveal effects of language on color discrimination. Proceedings of the National Academy of Sciences, 104(19), 7780–7785.
    https://doi.org/10.1073/pnas.0701644104 · PNAS
  14. Chen, A. Y., Halper, P., & Afshordi, N. (2025). Copenhagen survey on black holes and fundamental physics. arXiv preprint arXiv:2503.15776.
    https://arxiv.org/abs/2503.15776
  15. Liu, R., Quintin, J., & Afshordi, N. (2026). Ultraviolet completion of the Big Bang in quadratic gravity. Physical Review Letters, 136(11), 111501.
    https://doi.org/10.1103/6gtx-j455 · APS Journals
  16. Iskandar, G. R., Salim, F., Jihan Azmi Kusmawan, & Rahmadsyah Rangkuti. (2025). Linguistic relativity in Gap × Katseye 2025 Instagram advertisement: A Whorfian weak hypothesis approach. Jurnal Penelitian Inovatif, 5(4), 3181–3196.
    https://doi.org/10.54082/jupin.1947
  17. Sellars, W. S. (1962). Philosophy and the scientific image of man (R. Colodny, Ed.). PhilPapers; Humanities Press/Ridgeview.
    https://philpapers.org/rec/SELPAT
  18. Vacura, M. (2015). Sellars's concept of manifest and scientific image. E-LOGOS: Electronic Journal for Philosophy, 22(2), 94–102.
    https://doi.org/10.18267/j.e-logos.426 · E-LOGOS
  19. Pinto, A. O. (2024). Language and thought in Wittgenstein. Advances in Social Sciences Research Journal, 11(12), 14–23.
    https://doi.org/10.14738/assrj.1112.17871 · Scholar Publishing
  20. Wittgenstein, L. (1922). Tractatus logico-philosophicus (C. K. Ogden, Trans.). Kegan Paul, Trench, Trubner.
    https://www.gutenberg.org/files/5740/5740-pdf.pdf
  21. Wittgenstein, L. (1953). Philosophical investigations (G. E. M. Anscombe, Trans.). Blackwell.
    https://archive.org/details/philosophicalinv0000witt
  22. Polanyi, M. (1966). The tacit dimension. Doubleday.
    https://archive.org/details/tacitdimension00pola
  23. Kuhn, T. S. (1962). The structure of scientific revolutions. University of Chicago Press.
    https://archive.org/details/thomas-s.-kuhn-the-structure-of-scientific-revolutions
  24. French, S. (2020). Imagination in scientific practice. European Journal for Philosophy of Science, 10(3), 1–21.
    https://doi.org/10.1007/s13194-020-00291-z · Springer
  25. Norton, J. D. (2004). Einstein's investigations of Galilean covariant electrodynamics prior to 1905. Archive for History of Exact Sciences, 59(1), 45–105.
    https://doi.org/10.1007/s00407-004-0085-6 · Springer
  26. Wigner, E. P. (1960). The unreasonable effectiveness of mathematics in the natural sciences. Communications in Pure and Applied Mathematics, 13(1), 1–14.
    https://doi.org/10.1002/cpa.3160130102 · Caltech
  27. Rovelli, C. (1996). Relational quantum mechanics. International Journal of Theoretical Physics, 35(8), 1637–1678.
    https://doi.org/10.1007/BF02302261 · Springer
  28. Fuchs, C. A., Mermin, N. D., & Schack, R. (2014). An introduction to QBism with an application to the locality of quantum mechanics. American Journal of Physics, 82(8), 749–754.
    https://doi.org/10.1119/1.4874855 · AIP Publishing
  29. Zinkernagel, H. (2016). Niels Bohr on the wave function and the classical/quantum divide. Studies in History and Philosophy of Modern Physics, 53, 9–19.
    https://doi.org/10.1016/j.shpsb.2015.11.001 · ScienceDirect
  30. Li, J. (2024). A historical analysis of the independent development of calculus by Newton and Leibniz. Theoretical and Natural Science, 30, 1–5.
    https://doi.org/10.54254/2753-8818/30/20240739
  31. Jevtić, F. D., Kostić, J., & Maksimović, K. (2024). Reflecting on beauty: The aesthetics of mathematical discovery. arXiv preprint arXiv:2405.05379.
    https://arxiv.org/abs/2405.05379
  32. Ritson, S., & Camilleri, K. (2015). Contested boundaries: The string theory debates and ideologies of science. Perspectives on Science, 23(2), 192–227.
    https://doi.org/10.1162/POSC_a_00168 · MIT Press
  33. Gasparinetti, S. (2025). Ontologies for quantum field theory. Philosophy Compass, e70061.
    https://doi.org/10.1111/phc3.70061 · Wiley
  34. Torretti, R. (1978). Philosophy of geometry from Riemann to Poincaré. D. Reidel.
    Springer · PhilPapers
  35. French, S., & Ladyman, J. (2003). Remodelling structural realism: Quantum physics and the metaphysics of structure. Synthese, 136(1), 31–56.
    https://doi.org/10.1023/A:1024156116636 · PhilPapers
  36. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. University of Chicago Press.
    University of Chicago Press
  37. Lakoff, G., & Núñez, R. E. (2018). Where mathematics comes from: How the embodied mind brings mathematics into being. PhilPapers.
    https://philpapers.org/rec/LAKWMC
  38. Hossenfelder, S., & McGaugh, S. S. (2018). Is dark matter real? Scientific American, 319(2), 36–43.
    https://doi.org/10.1038/scientificamerican0818-36 · PubMed
  39. Herdeiro, C. A. R., & Lemos, J. P. S. (2019). The black hole fifty years after: Genesis of the name. arXiv preprint arXiv:1811.06587.
    https://arxiv.org/abs/1811.06587
  40. Kosslyn, S. M., Ganis, G., & Thompson, W. L. (2001). Neural foundations of imagery. Nature Reviews Neuroscience, 2(9), 635–642.
    https://doi.org/10.1038/35090055 · Nature
  41. Bohm, D. (1980). Wholeness and the implicate order. Routledge.
    https://doi.org/10.4324/9780203995150 · Internet Archive
  42. Cassini, A., & Levinas, M. L. (2019). Einstein's reinterpretation of the Fizeau experiment: How it turned out to be crucial for special relativity. Studies in History and Philosophy of Modern Physics, 65, 55–72.
    https://doi.org/10.1016/j.shpsb.2018.08.004 · ScienceDirect
Labels: Science and Philosophy Language Cosmology Linguistics