Abstract
This article extends the theory of symbolic fit by distinguishing a third axis, cosmic fit, from the two already established, mesocosmic fit and intersymbolic fit. Cosmic fit names the answerability of a symbolic articulation to the cosmos, considered independently of the particular organization mesocosmic coordination imposes upon it. Mesocosmic fit names the relation between an articulation and the human-scale world sustained through the five mediations, and possesses a genuine internal coherence of its own because those five mediations can cohere with, or conflict with, one another. Intersymbolic fit names the relation among symbolizations themselves, and it is intersymbolic fit, not a separate cosmic coherence, that accounts for the convergence among independently generated articulations that gives confidence in a cosmic claim. The three axes are partially independent: success along one guarantees nothing along another. The argument proceeds through three contrasting cases, a fabric swatch, a Genesis panel in Berlin that turns out to copy a lost painting by Hieronymus Bosch, and dark matter, moving from dense mesocosmic availability toward almost total mesocosmic inaccessibility, before returning to the general architecture the three cases reveal together.
I. The Hidden Ambiguity of Fit
Begin with an apparently trivial question. What does it mean for a symbolic articulation to fit? The word appears deceptively simple, appearing across an implausibly wide range of uses without apparent strain. A description fits an object. A portrait fits its sitter. A map fits a landscape. A mathematical model fits a set of observations. A theory fits the evidence gathered in its support. A fabric sample fits the sofa a customer has in mind. A biblical painting fits scripture and the doctrinal tradition surrounding it. A cosmological model fits astronomical measurement to many decimal places. Each of these sentences sounds perfectly ordinary, and each invites the same unexamined assumption, that fit names one relation, holding between a symbolic articulation and whatever that articulation is supposed to be about.
The philosophical temptation has been to honour that assumption by building theories of representation, correspondence, reference, denotation, resemblance, or exemplification around a single, if internally sophisticated, relation. However these theories differ among themselves, their underlying architecture remains approximately dyadic, symbolic articulation on one side, world on the other, with a single question asked between them, whether the one has captured the other successfully. This article’s contention is that the dyadic architecture is not simply incomplete. It mistakes what fit actually is, because the same symbolic articulation can stand in several relations of fit simultaneously, relations that vary independently of one another and that a single dyad cannot register.
Introduce the three terms without yet fully theorizing them. Cosmic fit names the answerability of a symbolic articulation to the realities and relations of the cosmos, a relation that holds independently of whatever organization human mesocosmic coordination happens to impose on those realities, and that therefore cannot be reserved for the microscopic or the astronomical alone. Mesocosmic fit names the relation between an articulation and the human-scale world in which recursive beings actually coordinate, constituted, as Living Value Theory’s Core Principles already establish, through the five mediations of embodiment, being-with, dwelling, multimateriality, and multisymbolization. Intersymbolic fit names the relation among symbolic articulations themselves, already developed at length elsewhere in this corpus as the relation through which one symbolization renders, translates, compresses, or expands the differentiations stabilized within another.
An example makes the independence vivid before any of the three terms has been properly theorized. Dark matter, discussed at length below, possesses very high mesocosmic intelligibility, because matter is one of the most ordinary categories available to embodied human coordination, something that has heft, that occupies space, that can in principle be handled. It coordinates an extraordinarily successful body of astronomical and mathematical symbolization, giving it very high intersymbolic fit as well. And yet what cosmic reality actually corresponds to the phrase remains, at the time of writing, unresolved. An equation from quantum mechanics presents close to the reverse profile, extraordinary cosmic fit, tracking relations among subatomic phenomena with a precision nothing else can match, while remaining close to untranslatable into anything a body, walking through a room, could intuitively grasp. Neither case is anomalous. Both simply demonstrate that the three relations cannot be collapsed into one another.
State the major proposition early, because the remainder of the article exists only to work out its consequences. Symbolic fit is multidimensional. A symbolic articulation does not stand in a single relation to a single world. It stands, at once, in relations of fit to the cosmos, to the mesocosm, and to other symbolizations, and these relations can reinforce, compensate for, conceal, or contradict one another without any one of them determining the others. What follows asks what changes once this apparently modest distinction is taken seriously, moving deliberately from the most ordinary case a theory of representation could choose to the most remote.
II. Why Representation Has Looked Like One Problem
Understanding why the three relations were so often treated as one requires a brief diagnostic reconstruction rather than a history of theories of representation, since the purpose here is not exhaustive but explanatory.
Begin with the extraordinary persuasive force of mesocosmic examples. This is a cat. This table is brown. This sample is the colour I want. The picture depicts a horse. Sentences of this kind make representation appear to involve nothing more than a symbol and an independently available thing, because the thing in question is already perceptually present to speaker and listener alike. Both inhabit approximately the same spatial environment. Both possess sensory capacities differentiated in similar ways. Both have been inducted, through years of embodied coordination, into shared practices of attending, pointing, handling, and classifying. None of this achievement is symbolic in the first instance. It is mesocosmic, sedimented across embodiment and being-with long before any word is spoken, and it works so well that it disappears from view entirely, leaving only the appearance of a symbol standing directly before a thing.
This is the philosophical illusion of direct reference, and it is an illusion generated by success rather than by error. The enormous background achievement of mesocosmic fit vanishes precisely because mesocosmic fit, when it is working, is exactly the kind of coordination that requires no further attention, in keeping with the principle, established elsewhere in this corpus, that successful coordination mutes itself into invisibility. A theory of representation built entirely from such cases inherits the invisibility along with the apparent simplicity, and concludes that representation is a single relation because the relation it is actually observing, mesocosmic fit operating at very high efficiency, has already done nearly all of the work before any explicit symbolic claim is made.
Ernst Cassirer deserves credit here, though a narrow and specific one. His recognition that human symbolic activity is heterogeneous, that language, art, myth, mathematics, and science cannot be treated as variants of a single symbolic medium, was an important advance against exactly the collapse this article is concerned with. But Cassirer’s heterogeneity thesis stops one step short of what is needed. Recognizing that different symbolic systems exist is not the same achievement as theorizing the relations of fit that hold among them, or the relations each bears to the mesocosm and to the cosmos beyond it. Cassirer supplies a taxonomy of symbolic forms. He does not supply an account of how those forms coordinate, fail to coordinate, or translate into one another, which is precisely the question intersymbolic fit exists to answer.
Nelson Goodman is the far more productive interlocutor, because his rejection of naive resemblance theories, together with his sophisticated distinctions among denotation, exemplification, and the differing symbolic systems within which each operates, comes remarkably close to the present problem without quite arriving at it. Goodman’s example of the fabric sample, a customer presenting an upholsterer with a swatch and saying, in effect, I want the sofa like this, has become a classic illustration of exemplification, certain properties of the sample being symbolically selected for projection onto the finished object, most others being excluded. It is worth reconstructing this example in full, because it looks philosophically simple only because almost everything difficult about it has already been solved, invisibly, by the mesocosm. The sofa is where the argument of this article properly begins.
III. The Sofa: The Supposedly Easy Case
Reconstruct Goodman’s fabric sample carefully, because the case is deceptively rich. A person presents a swatch of fabric to an upholsterer and says, I want the sofa like this. What does this denote?
The swatch, considered simply as a physical object, has innumerable properties. It has a particular size and a rectangular shape. It has mass, temperature, age, and a characteristic smell. It has a specific weave, a specific pattern, a specific chemical composition, perhaps a frayed edge, and it is, at the moment of utterance, being held at a particular angle to the light. Nobody supposes the customer is requesting a sofa three inches square with one frayed edge. Some of the swatch’s properties are being symbolically selected for projection onto the sofa that does not yet exist, and the great majority are being excluded, and the excluding is doing at least as much work as the selecting.
Goodman’s own concept of exemplification captures part of this achievement, since it explains how a sample can refer by possessing, rather than merely denoting, the properties it projects. But the achievement is not exhausted by exemplification alone, because the sample does not symbolize on its own. Several kinds of coordination are operating together at the moment the swatch is held out and the sentence spoken. The fabric is materially present, and can be seen, and can be touched. The pointing gesture establishes what, among everything else in the room, is relevant to attend to. Ordinary language supplies the phrase like this, which does grammatical work no swatch could do unaided. Both parties share an understanding that this is a transaction concerning upholstery rather than, say, a gift or a specimen. Both bring prior, embodied knowledge of what sofas are and what kinds of properties typically survive the journey from sample to finished object. And the anticipated future sofa itself, not yet existing anywhere, nonetheless supplies the absent target the entire transaction is organized around.
The symbolic achievement of the fabric sample is therefore distributed across a dense field of coordination rather than located in the sample alone, and this is the crucial move the case makes possible. The sofa is not merely an instance of mesocosmic fit. It is a case in which an unusually dense mesocosmic background makes an intersymbolic achievement, the coordination between a swatch, a gesture, a sentence, and a future object, look effortless rather than remarkable. The stronger proposition follows directly. Mesocosmic fit can lower the burden that would otherwise fall on intersymbolic fit. Where participants already share an extraordinarily thick background of embodied, relational, and practical coordination, comparatively little symbolic labour is required to achieve a result that would, in the absence of that background, require elaborate specification.
This deserves to be stated as a general theorem rather than left as an observation local to upholstery. Mesocosmic availability does not eliminate intersymbolic fit. It makes intersymbolic fit so easy that it becomes invisible. The philosophical tradition has typically treated cases like this one, ostensive cases where a symbol seems finally to touch the world directly, as the bedrock on which more complex representational achievements are built. The sofa shows why this gets the order of dependence backwards. The gesture, the utterance, the sample, the visual attention of both parties, and the shared classification of the transaction as a purchase of upholstery must already fit one another before anything can be pointed at successfully. What looks like symbolism finally reaching the world is, on inspection, already an achieved coordination among several symbolic and mesocosmic mediations, coordination that mesocosmic density has simply made frictionless enough to seem immediate. Ostension is not the primitive case intersymbolic fit is built up from. It is one of intersymbolic fit’s most thoroughly disguised achievements.
The transaction does not end with the utterance. When the finished sofa arrives, the original symbolic specification is tested through an entirely different register of coordination. The customer sees the upholstery under the room’s actual light rather than the shop’s. She touches it, sits on it, notices how it resists or yields to the body’s weight, perhaps catches an unexpected smell from the fabric, and places the whole object within a room it must now fit in the most literal, spatial sense. What was specified symbolically becomes recursively evaluated against a mesocosmic object that now exists in its own right, capable of surprising the very expectations that were supposed to have determined it in advance. This is not a simple relation between a word and a world it labels. It is a complex, distributed achievement of fit across several kinds of coordination at once, most of them not linguistic in the first place, and the achievement remains visible for as long as the sofa continues to be evaluated rather than simply used. Once it is simply used, once it has been muted into the unremarkable furniture of a room, the achievement disappears from view in exactly the way mesocosmic fit tends to disappear once it has succeeded.
IV. Mesocosmic Fit and the Five Mediations
The mesocosm, as established elsewhere in this corpus, is not simply the world at a convenient physical scale. It is the recursively inhabitable world within which living beings coordinate, act, and evaluate the success of their own coordination. Mesocosmic fit can now be defined formally as the relation between a symbolic articulation and this world, and because the mesocosm is constituted through five irreducible mediations, mesocosmic fit is never a single, homogeneous relation. It is an achievement of greater or lesser coherence across all five at once.
The sofa makes each mediation legible in turn. Multisensorial embodiment registers the sofa visually, tactilely, proprioceptively, and often olfactorily and acoustically as well, and these registrations need not agree with one another. Upholstery can look soft and prove firm under the body’s weight. A finish can look natural and smell unmistakably synthetic. Being-with places the sofa within a field of interpersonal negotiation, since people routinely disagree about comfort, beauty, appropriate use, and rightful ownership, and the original transaction with the upholsterer was itself an act of interpersonal coordination before it was anything else. Multimaterial forming concerns what the sofa affords and resists, sitting, reclining, cleaning, carrying, repairing, rearranging. Multiversal dwelling concerns where the sofa belongs, since its size, orientation, and function must fit a room, a household, and an entire pattern of dwelling, and a sofa that cannot pass through the doorway demonstrates spectacular mesocosmic misfit however accurate every symbolic specification of it happened to be. Multisymbolism, finally, is the mediation through which the sofa is articulated at all, through the word sofa, through measurements, through fabric codes, through prices, diagrams, instructions, advertisements, and the original swatch that started the whole transaction.
Because the five mediations are irreducible to one another, they can cohere, and they can just as easily conflict, and several compact examples make the pattern visible. A sofa can be visually and multisymbolically valued as beautiful while remaining, in embodied terms, uncomfortable to sit in for any length of time, a conflict between multisymbolism and multisensorial embodiment. A sofa can be comfortable for the individual body while proving, in the register of being-with, socially unacceptable within a household that finds its style or its placement objectionable. A sofa’s measurements can be symbolically correct to the centimetre while its actual passage into the room, a matter of multimaterial forming set against multiversal dwelling, proves physically impossible. A sofa advertised as natural linen through its accompanying multisymbolic apparatus can be experienced, on contact, as unmistakably synthetic, a direct conflict between what the label claims and what the body reports.
These examples can be generalized without needing to be multiplied further. Because the five mediations are independent sources of differentiation, mesocosmic coherence is better understood as a network of pairwise relations among them than as a simple checklist of five items each separately satisfied or not. Being-with can fit or fail to fit multisensorial embodiment, as when a household’s shared taste conflicts with what an individual body actually finds comfortable to sit in. Multiversal dwelling can fit or fail to fit multimaterial forming, as the doorway already showed. Multisymbolism can affirm or contradict embodiment directly, as the label claiming natural linen against a synthetic touch already demonstrated. Five mediations generate ten such pairings, and no symbolic specification, however careful, can guarantee that coherence in one pairing carries over to the others. Mesocosmic fit is therefore relational all the way down, and this is precisely what allows it to fail locally, along one or two pairings, while remaining broadly intact elsewhere, which is closer to how mesocosmic misfit is actually experienced than any single aggregate verdict of fit or misfit could capture.
The methodological point generalizes well beyond furniture. Mesocosmic fit is itself internally structured rather than a single undifferentiated relationship between an organism and its environment. It is an achievement of greater or lesser coherence across five mediations that answer to genuinely different demands, and that no single symbolic specification, however precise, can guarantee will align. This gives the theory of symbolic fit a model it will use repeatedly for the remaining two kinds of fit as well, without multiplying the number of fundamental fits beyond the three the article began with. Just as mesocosmic fit can succeed along some mediations while failing along others, intersymbolic fit, taken up properly in the section that follows, can succeed between some symbolic systems while failing between others. Cosmic fit, introduced later still, will turn out to work differently again, answerable to the cosmos directly rather than structured by internal parts of its own, which is exactly why section nine will need to distinguish it carefully from the other two rather than assuming every kind of fit possesses internal coherence in the same sense. The five mediations are therefore not merely a description of mesocosmic fit specifically. They are the first demonstration, within this article, that a kind of fit can be simultaneously singular in name and plural in structure, a lesson every subsequent section will need.
V. Cranach: Representing What Nobody Could Inspect
Move now from the easy case to a harder one, remaining within the visual arts but leaving behind anything a customer could hold in her hand.
Lucas Cranach the Elder painted numerous depictions of the Fall, and one of them, a panel now in the Gemäldegalerie in Berlin, forms part of a triptych of the Last Judgement modelled closely on a composition by Hieronymus Bosch. The painter faces a representational problem radically unlike the upholsterer’s. There is no available Eden to inspect. There is no serpent to observe. There is no historical likeness of Eve against which the finished panel could be checked. If representation is understood, as the dyadic architecture criticized in section two would have it, primarily as fit between a symbol and a perceptually available target, biblical painting of this kind becomes philosophically puzzling almost immediately, because the target was never available to begin with, to Cranach or to anyone else who ever painted it.
And yet the painting is not arbitrary. It is tightly constrained, and the constraint comes from elsewhere. The relevant symbolic systems include the biblical text itself, centuries of theological interpretation built up around it, an existing pictorial tradition of Genesis imagery stretching back well before Cranach, a set of iconographic conventions governing how temptation, nakedness, and sin were to be shown, the ecclesiastical and liturgical contexts within which such paintings were displayed and read, and the ordinary-language understandings, current among Cranach’s own public, of what bodies, snakes, fruit, and gendered temptation were supposed to look like. The painting must achieve fit with this entire field of prior symbolization rather than with any perceptually available original. This does not make the painting any less mesocosmic. A panel is a physical object, encountered by an embodied viewer, hung within an ecclesiastical space, and every one of the five mediations is at work in its production and reception exactly as each was at work in the sofa’s. What differs is that the painting’s cosmic target, whatever actually happened, if anything, in a garden nobody has ever stood in, cannot adjudicate the painting’s symbolic success the way the finished sofa could adjudicate the swatch’s. With cosmic adjudication unavailable, intersymbolic fit becomes the register in which the painting’s success is actually judged, and becomes, for exactly that reason, unusually visible rather than unusually dominant.
The serpent in the Berlin panel makes this with unusual clarity. It climbs the tree above Adam and Eve bearing a woman’s head, a woman’s hair, and a woman’s torso, grafted onto a scaled, taloned, unmistakably reptilian lower body. Genesis supplies no visual specification of the serpent’s face. Cranach’s decision to give it Eve’s own features belongs to a wider iconographic convention, current across Northern European Genesis imagery of the period, in which the tempter is rendered with a woman’s likeness precisely in order to bind the serpent visually to the very party it is about to seduce. The image becomes interpretable only through its relations to scripture, to accumulated theological reflection on the nature of temptation, and to a pictorial tradition already primed to receive exactly this kind of visual pun. None of this could be checked against Eden. All of it could be checked, and continually was checked, against the surrounding field of symbolic articulation the painting had to answer to.
Mesocosmic fit has not disappeared from the case, however, and this matters as much as its diminished role. Eve must still look bodily possible. The fruit must appear graspable by an actual hand. The tree occupies real pictorial space, its trunk thick enough to bear the weight the composition asks of it. Bodies retain orientation, proportion, and scale. Gestures, Adam reaching upward for a second fruit even as Eve’s hand still rests against him, remain recognizable as gestures a body could actually perform. The unavailable biblical world is rendered, that is, through mesocosmically available forms, even while the content those forms are being asked to convey answers overwhelmingly to intersymbolic rather than mesocosmic demands.
A further layer complicates the case in a way the sofa could never have anticipated, and it deserves stating plainly rather than left as a footnote. The Berlin panel is itself understood by specialists, and recorded as such in the panel’s own catalogue entry, to be a copy, made during or after Cranach’s documented visit to the Netherlands in the first decade of the sixteenth century, of a Bosch composition that has since been lost outright. The lost original survives nowhere as an object. It survives only indirectly, through Cranach’s Berlin panel on one hand and, independently, through a later print by Hieronymus Cock that records substantially the same composition on the other. Neither witness can be checked against the painting it derives from, because that painting no longer exists to check anything against. What confidence there is in reconstructing the lost Bosch composition rests entirely on the convergence of two independent symbolic chains arriving, by different routes, at compatible articulations of something neither of them can any longer point to directly.
This is worth naming for what it structurally is, because the case to come, several sections later, will turn out to share its architecture exactly. A panel in Berlin depicting a scene nobody could ever have witnessed is already, on inspection, a case of intersymbolic convergence substituting for an original that mesocosmic fit could never have supplied even in principle, since the original was itself lost before any twentieth or twenty-first century viewer had the chance to stand before it. The key contrast for now can be stated simply. The sofa is mesocosmically anchored, and this is precisely what allows its considerable intersymbolic achievement to be overlooked. Cranach’s Eden, doubly unavailable, first as biblical scene and then, in this particular case, as a lost painting behind the painting that survives, exposes intersymbolic fit for what it has been doing all along.
VI. Cosmic Fit: When Symbolic Systems Leave the Mesocosm
Cosmic fit can now be introduced in its own right, and the temptation to define it merely as fit with astronomical fact, or to reserve it for the very small and the very distant, should be resisted from the outset. Cosmic fit is not a region of reality where a special kind of fit happens to apply. It is an aspect of the relation between a symbolic articulation and reality as such, the aspect that asks whether the articulation answers adequately to what is the case independently of the particular organization mesocosmic coordination has imposed upon it. Mesocosmic fit asks a different question of the very same articulation, whether it works within the recursively inhabitable world sustained through the five mediations. The two questions are logically independent, and, crucially, the same articulation can be assessed by both at once. The sentence this sofa is two metres long can possess excellent mesocosmic fit, coordinating comfortably with embodiment, dwelling, and the room it must pass through, while also possessing excellent cosmic fit, since two metres names a length that holds regardless of any human coordination organized around it. Quantum mechanics presents the more instructive asymmetry, extraordinary cosmic fit paired with mesocosmic fit that remains, for nearly everyone who has not spent years training to read it, close to unintelligible.
What a handful of reserved domains, a protein interaction beneath perceptual threshold, a quantum state, deep geological time, cosmological expansion, actually illustrate is therefore not that cosmic fit belongs to them specifically, but that they are domains where mesocosmic fit becomes close to unavailable, leaving cosmic fit to operate almost entirely on its own, unsupported by the embodied intelligibility that quietly assists it everywhere mesocosmic fit is also strong. This is also how cosmic fit differs from dwelling as already established in this corpus. Dwelling is the found, nonrecursive orientation within which every living being coordinates, terrain, weather, the solar cycle, the great astronomical regularities that were never made by anyone and yet remain perceptible, trackable, and available to embodied coordination across a human lifetime. Cosmic fit is not simply dwelling extended further outward. It is a question that can be asked of dwelling itself, since a mountain’s found stability is answerable to cosmic fact, geological and gravitational, quite apart from whatever orientation it happens to afford a climbing body, and it is a question that can equally be asked of a quantum state that no orientation, found or made, could ever afford at all.
What makes cosmic fit achievable in domains where mesocosmic support has fallen away is one of the more remarkable capacities of specialized symbolic systems. Mathematics, statistical notation, spectroscopy, chemical notation, genomic sequencing, computational modelling, and instrumentally mediated measurement each possess a freedom that ordinary language largely lacks, the freedom not to resemble anything a body ordinarily encounters. A wavefunction need not look like anything. A genomic sequence need not sound like anything spoken. These systems can introduce as many decimal places as a measurement demands, define new variables purpose-built for a single relation, construct wholly new notational conventions, and represent relationships that possess no ordinary-language analogue whatsoever, because nothing in their construction requires them to remain intelligible at the scale of an embodied, dwelling, being-with creature.
This is precisely why L3 differentiation matters so much wherever cosmic fit must operate without the mesocosmic support available elsewhere. At L3, the level at which something becomes symbolically articulated as this particular thing embedded within an ongoing coordinative field, specialized systems can push differentiation to an extraordinarily fine grain, well beyond anything a mesocosmically anchored articulation would ever need or tolerate. The central proposition follows directly from this observation. L3 symbolic systems can increase cosmic fit precisely by decreasing their dependence on mesocosmic fit, trading intuitive resemblance to lived experience for a precision the cosmos, at scales the mesocosm cannot itself register, actually requires. This should not be read as an unfortunate cost scientific symbolization happens to incur. It is one of the constitutive powers poietic symbolization makes available, the capacity, unique among the theory’s five mediations, to deposit differentiations so exact that they can track a reality no living body will ever directly perceive.
VII. Dark Matter: The Case with No Swatch
Introduce dark matter now as the limiting case against which Goodman’s sofa can be read in reverse. There is no dark matter sample. Nobody can point at it, touch it, or photograph it in any ordinary sense. Nobody has isolated a confirmed dark matter particle in a laboratory. And yet dark matter functions as one of the central entities of contemporary cosmology, load-bearing across an enormous range of astrophysical explanation. What kind of fit makes an entity like this scientifically compelling in the total absence of anything resembling Goodman’s swatch?
The symbolic ecology surrounding dark matter is worth reconstructing briefly, not exhaustively, since a fuller history belongs elsewhere in this corpus. Fritz Zwicky’s early measurements of galaxy cluster velocities, interpreted through the mathematics of the virial theorem, first suggested that visible luminosity accounted for far too little mass to hold clusters gravitationally together. Later, galactic rotation curves, obtained through optical spectroscopy and eventually corroborated through radio observation of neutral hydrogen at twenty one centimetres, showed stars and gas orbiting far too fast at a galaxy’s outer edge for the visible matter alone to explain. Gravitational lensing supplied an independent route again, reconstructing mass distributions from the way background light bends around foreground structure. Measurements of the cosmic microwave background, expressed through its temperature and polarization and reduced to an angular power spectrum, supplied a route independent of all the others still. Large scale structure surveys, statistical inference, and increasingly sophisticated numerical cosmological simulations added further, largely independent lines. Particle detection experiments, so far unsuccessful in isolating a candidate directly, added a further constraint from below.
Data is a poor word for this heterogeneity, and it is worth stressing why. A radio wave is not itself data. It is a physical event, a fluctuation in an electromagnetic field passing through an antenna. Instrumentation transduces that event into an electrical signal. Processing calibrates the signal against known standards. Symbolization then renders the calibrated result as a numerical value, a spectrum, a curve, or a map. Scientists themselves call several of these intermediate products data, raw data, detector data, calibrated data, and nothing in the argument here depends on policing that usage or insisting on one privileged stage at which data properly begins. The point is different, and, on reflection, stronger. Data names a family of heterogeneous products generated at very different stages of material transduction and symbolic stabilization, and precisely because the family is this heterogeneous, data cannot function as a theoretically primitive epistemological category, a single kind of thing scientific epistemology could simply accumulate more of. The scientific object dark matter has become emerges through an entire chain of transformation among heterogeneous symbolic systems, from physical event, through instrumental transduction, through calibrated numerical articulation, through spectrum, through inferred velocity, through rotation curve, through inferred mass distribution, through statistical comparison, to cosmological model, and calling every stage of this chain simply data flattens distinctions that matter enormously to how much confidence any particular stage deserves.
Dark matter becomes scientifically compelling, in short, because heterogeneous symbolic systems, arrived at through independent instrumental and mathematical routes, converge on compatible articulations of something none of them can inspect directly. This licenses a major proposition about scientific epistemology generally. Scientific evidence frequently derives its force from intersymbolic convergence rather than from the simple accumulation of observations within a single symbolic register. The sheer amount of evidence matters, undeniably, but the independence and heterogeneity of the symbolic pathways generating that evidence matters differently, and arguably matters more. Ten thousand additional rotation curve measurements, however carefully gathered, remain epistemically continuous with the rotation curve evidence already available. The emergence of gravitational lensing evidence, generated through an entirely different symbolic and material chain, is not continuous with rotation curve evidence in the same way, and its independent convergence on a compatible answer carries a distinct evidential weight that mere accumulation within one chain could never supply on its own.
This is why triangulation, a word scientists reach for reflexively, needs a more precise theoretical reconstruction than it usually receives. Scientific triangulation is best understood as a particularly disciplined practice of establishing intersymbolic fit across partially independent routes, testing, and where it succeeds producing, convergence among symbolic pathways bearing on a reality none of them can access directly. This leaves room for triangulation to operate usefully even among closely related symbolic systems, rather than requiring every instance of it to draw on maximally heterogeneous pathways before it counts as triangulation at all. The Cranach panel discussed above, itself a copy of a Bosch composition lost beyond recovery, rested its own reconstructive confidence on exactly this architecture in miniature, two independent symbolic witnesses, a painted copy and a printed record, converging on a composition neither witness could any longer check against an original. Dark matter simply runs the same architecture at civilizational scale, with dozens of independent symbolic pathways rather than two, and with a cosmic rather than an art-historical target standing, permanently, just beyond what any of them can inspect directly.
VIII. Why Data Is the Wrong Unit of Scientific Epistemology
The dark matter case generalizes into a claim about scientific epistemology that reaches well beyond astrophysics. Scientific discourse treats data as though it named a single, epistemologically homogeneous category, but what gets called data in practice includes images, numerical measurements, genomic sequences, spectra, maps, physical traces, interview transcripts, statistical counts, and reconstructed probability distributions, and these are not interchangeable symbolic products differing only in subject matter. They are differently constituted symbolizations, each preserving certain distinctions with precision while remaining silent, or actively lossy, on others, and each capable of being transformed into the others only at a cost that a flat appeal to data conceals rather than reveals.
Return to the dark matter chain to see the point concretely, though briefly, since the fuller reconstruction belongs to the previous section. An electromagnetic event becomes an instrumental signal, becomes a calibrated numerical articulation, becomes a spectrum, becomes an inferred velocity, becomes a rotation curve, becomes an inferred mass distribution, becomes a statistical comparison, becomes a cosmological model. Calling every one of these a further piece of data erases the distinct operations actually occurring at each step, which are better distinguished as physical interaction, instrumental transduction, symbolic articulation, transformation between symbolic systems, and eventual stabilization into a form durable enough to be cited, taught, and challenged by others. Much of what scientists actually spend their working lives doing is not, on this reconstruction, collecting more data in any simple sense. It is achieving transformations among symbolizations that manage not to lose the distinctions that matter for the question at hand, an achievement this corpus has elsewhere analysed, in an institutional register, as intersymbolization, together with its characteristic hazards of compression, expansion, and irreducible remainder.
The same hazards apply here, restated for a scientific rather than an institutional setting. A rotation curve compresses an enormous amount of raw spectroscopic detail into a single tractable form, and the compression, as in every case of intersymbolic compression, is not inherently damaging, since no cosmological inference at any useful scale could proceed without it. A cosmological model expands a handful of parameters into an entire predictive apparatus, and the expansion, again as elsewhere, risks being mistaken for discovery when it is in fact production, the sophistication of the model demonstrating the resources of the symbolic system doing the modelling rather than automatically demonstrating anything about the reality being modelled. And every transformation in the chain, however carefully executed, produces its own remainder, whatever fails to survive the passage from one symbolic register into the next, a possibility no accumulation of further measurement within a single register can by itself repair.
Scientific robustness, properly understood, arises from two distinct sources that should not be collapsed into one. Repeated articulation within a single symbolic chain, more measurements along the same rotation curve, more spectra of the same kind, can genuinely increase precision and reliability, and nothing in this argument denies it. But independently generated chains converging on compatible articulations despite passing through entirely different transformations, different instruments, different assumptions, different symbolic architectures, add a further and distinct kind of warrant, because what survives that convergence has survived routes that could, in principle, have diverged and did not. Both matter, and they are not competing for the same epistemic credit. This reconstruction supplies a considerably more precise account of evidential triangulation than the word usually receives in scientific discourse. It also licenses the stronger epistemological proposition this section exists to state plainly. The fundamental unit of scientific epistemology is, more often than the vocabulary of data admits, not the datum at all but the relation among symbolic articulations, precisely the relation this article has been calling intersymbolic fit throughout. This is not offered as a complete philosophy of science, only as a necessary preparation for the sections that follow, so that the general theory of fit developed here does not collapse, in its own account of scientific knowledge, into exactly the flattened vocabulary it has spent this section criticizing.
IX. Internal Coherence and Cross-Fit
Return now to the general architecture and draw its remaining consequences together. It is tempting, having distinguished three kinds of fit, to assume they must relate to internal structure in the same way, each possessing some internal coherence that parallels the other two. The temptation should be resisted, because it is not true, and giving in to it would quietly undo the distinction this article exists to defend.
Mesocosmic fit’s internal coherence is genuine and internal to the mesocosm itself, constituted, as section four showed, through the greater or lesser alignment of the five mediations with one another. Intersymbolic fit’s internal coherence is equally genuine and equally internal to what intersymbolic fit is, since intersymbolic fit just is a relation among symbolic systems, and asking how well several such systems cohere with, translate into, overlap with, or contradict one another is asking a question intersymbolic fit exists to answer. Cosmic fit is different in kind. Cosmic fit is the answerability of an articulation to cosmic reality, not a relation among articulations at all, and it therefore has no internal structure of its own to parallel the other two. When independently generated cosmically oriented articulations, mathematical, spectroscopic, statistical, computational, converge upon compatible descriptions of the same underlying reality, as the astronomical chains discussed above did in the case of dark matter, that convergence is intersymbolic fit operating across several routes at once, not a separate cosmic species of coherence standing alongside the other two. What convergence licenses is greater confidence that the cosmic fit of the converging articulations is good. Keeping this distinction sharp matters, because collapsing convergence into cosmic fit directly would quietly reintroduce the very reduction this article exists to resist, crediting cosmic fit with an achievement that belongs to intersymbolic fit throughout.
Beyond this, a further and looser question arises about fit among the three kinds of fit themselves, and several recurring configurations are worth naming, not as a rigid typology to be applied mechanically but as an analytic demonstration that the three fits are neither reducible to one another nor reliably correlated in either direction.
High mesocosmic fit can coexist with high intersymbolic fit while cosmic fit remains genuinely uncertain. Cranach’s Genesis panel illustrates the configuration precisely, bodily plausible, richly coherent with scripture and theological tradition and the wider pictorial corpus surrounding it, and entirely unaccountable, by its very nature, to any cosmic or historical fact about what actually occurred in an actual garden, if there ever was one. High cosmic fit can coexist with high intersymbolic fit while mesocosmic fit remains poor, the configuration much of mathematical physics approximates, its formalism coordinating spectacularly with instrument and observation while remaining close to unintelligible to an embodied, dwelling creature encountering it for the first time. High mesocosmic fit can coexist with high intersymbolic fit while cosmic fit is, in fact, poor, a more troubling configuration than either of the preceding two, since a cosmology or a diagnostic classification can be widely intelligible and institutionally coherent for generations while tracking the underlying reality badly, exactly the drift already analysed, in an institutional register, as the danger internal consistency among symbols can mask.
A further configuration deserves particular attention because it names a genuine hazard rather than a stable equilibrium. Cosmic fit can increase precisely as intersymbolic fit declines, and this is the configuration most likely to precipitate what is ordinarily called a scientific crisis, since a formalism gaining ever more exact purchase on cosmic relations can simultaneously lose its coherence with the surrounding body of established theory it was supposed to extend rather than overturn. And a final configuration, approximated by parts of quantum mechanics, pairs high intersymbolic fit, an extraordinary coordination between mathematics and experimental result, with mesocosmic fit that remains unresolved, since ordinary-language ontology, the question of what quantum formalism is actually about in terms a dwelling creature could inhabit, stays genuinely contested even where the mathematics itself is not.
None of these configurations should be read as a finished classification of scientific or symbolic possibility. Their purpose is narrower and more secure. They establish, analytically rather than merely by assertion, that the three kinds of fit vary independently of one another, that success along any one axis carries no automatic entailment for the other two, and that a theory of symbolization adequate to any of the cases this article has examined, the sofa, the panel in Berlin, dark matter, needs all three relations distinguished and none of them collapsed into either of the others.
X. Why Ordinary Language Keeps Bringing the Mesocosm Back
A further pattern deserves attention, offered here as a provisional hypothesis rather than a settled conclusion the rest of the article depends on. Specialized symbolic systems can travel extremely far from the mesocosm at L3, precisely as section six described, and yet scientific interpretation shows a remarkably consistent habit of returning to ordinary language once a formalism needs to be understood rather than merely operated. Equations eventually become statements about particles, fields, forces, strings, information, codes, networks, machinery, dark matter, and dark energy. Why should this be so, when the entire power of L3 specialization lay in its freedom from exactly this kind of mesocosmic resemblance?
One proposal is that L4 recursive stabilization, the level at which symbolic forms become abstracted and generalized into classifications, doctrines, or theories capable of travelling beyond the coordination that generated them, may be uniquely, or at least overwhelmingly, afforded by ordinary language rather than by specialized notation. Specialized symbolisms can perform extraordinarily sophisticated operations, including operations conducted upon their own prior outputs, and nothing said here diminishes that capacity. But L4 stabilization involves something categorically different from further specialized operation. It involves repositioning a symbolic articulation as an articulation, asking what kind of entity the articulation actually discloses, questioning whether the classification currently in use remains adequate, and settling the result into an account other minds, working in other registers, can inhabit and reason from in turn. Ordinary language appears to possess a distinctive capacity for exactly this kind of repositioning, because ordinary language itself emerged through, and remains saturated by, mesocosmic human existence in a way purpose-built mathematical or notational systems, however powerful within their own domain, do not.
If the proposal holds, L4 stabilization systematically reimports mesocosmic organization into domains from which L3 specialization had, quite deliberately, excluded it, and this supplies a structural rather than a merely rhetorical account of scientific metaphor. Dark matter becomes invisible stuff. Genes become codes. Cells contain machinery. Brains contain networks. Physics discovers fields, particles, and waves. Each of these borrows from mesocosmically available categories of objects, substances, movements, agents, containers, communication, and construction, categories embodiment, being-with, dwelling, multimateriality, and multisymbolism jointly make available to ordinary language long before any of them is pressed into technical service, without needing to be traced to any single mediation in particular. None of this implies that these concepts remain simple once adopted. Each can become highly technical, diverging enormously from anything its ordinary usage would suggest, to the point that a physicist’s field and a farmer’s field share almost nothing beyond the word itself. The claim under consideration concerns the route by which such concepts achieve L4 stabilization, not any continuing semantic identity with their everyday origins, and the distinction matters, because collapsing the two would recreate, in miniature, exactly the disguised L4 this corpus has warned against elsewhere, a technical abstraction mistaking its borrowed mesocosmic vocabulary for an unmediated description of what it actually names.
The hypothesis is deliberately left open to falsification rather than asserted as established fact. Whether mathematics can independently afford L4 stabilization without borrowing from ordinary language at all, whether formal logic can, whether computer code can, and whether an artificial symbolic system might eventually develop the capacity on terms no human mesocosm constrained, are questions this article sets aside rather than answers, reserving them for a later piece in this corpus better suited to taking them on directly.
XI. What This Changes about Representation
Return, finally, to the philosophical stakes raised at the outset. The classic dyad, representation on one side and represented reality on the other, has to be replaced by a genuinely relational architecture in which a single symbolic articulation can simultaneously fit, or fail to fit, the cosmos, the mesocosm through its five mediations, and the field of other symbolic articulations surrounding it. These three relations are not static once achieved. They change over time, sometimes reinforcing one another, sometimes compensating for one another’s weaknesses, sometimes concealing a weakness in one register behind an apparent strength in another, and sometimes coming into open conflict that no single act of symbolization can resolve on its own.
Revisit the three central cases in light of the completed architecture, and a progression emerges worth making explicit rather than left for the reader to notice unaided. The sofa shows the full architecture operating at its easiest, where an unusually dense mesocosmic coherence does almost all of the intersymbolic work invisibly. Cranach’s panel shows the architecture once the target cannot be inspected at all, exposing an intersymbolic achievement the sofa had let pass unnoticed, and the further discovery that the panel itself copies a now lost Bosch composition compounds the exposure, since even the intersymbolic field Cranach was answering to rests on a still earlier convergence nobody can any longer check directly. Dark matter shows the architecture at its most demanding, where even the ontological category of the target, the very question of what kind of thing dark matter is, has been generated through intersymbolic stabilization rather than discovered by inspecting anything directly, and where accountability to cosmic fact survives only through chains of instrumental mediation long enough that no single step in any of them could carry the argument alone. Nothing about this progression suggests that any of the three cases is representationally deficient relative to the others. Each achieves what its circumstances allow, through whichever combination of the three fits its circumstances make available, and a theory of representation that can only recognize the sofa’s kind of achievement will always find Cranach’s Eden, and dark matter still more, philosophically embarrassing rather than simply different in structure.
XII. Toward a General Theory of Symbolic Fit
Return, in closing, from the specifics of science and art to the larger problem this corpus exists to address. Living beings do not merely represent the worlds they inhabit. They continuously construct, test, maintain, destabilize, and reconstruct relations of fit between what they symbolize and everything that symbolization answers to, and the three kinds of fit distinguished in this article make it possible to analyse that ongoing labour without collapsing one relation into another or mistaking success in one register for success across all three.
Mesocosmic fit makes worlds inhabitable, sustaining the dense, embodied, relational, material, and symbolic coherence through which a living being can act without having to reconstruct its situation from nothing at every turn. Cosmic fit holds symbolic articulations answerable to realities that extend beyond the organization of human-scale existence altogether, reaching quantum states, deep time, and cosmological structure through specialized systems purpose-built for exactly that reach. Intersymbolic fit allows heterogeneous symbolic systems, none of them reducible to any master code, to coordinate with one another well enough that a civilization’s accumulated symbolic deposits can function as a single, if never perfectly unified, resource. None of the three is sufficient by itself. A symbolic world can be perfectly mesocosmically intelligible while being cosmically wrong, coherent to inhabit and false about the cosmos it claims to describe. And a scientific field can enter crisis precisely because its cosmic fit is improving, its formalism gaining ever sharper purchase on the relations it targets, while the intersymbolic relations that once held its community’s vocabulary together come under strain the improvement itself has caused.
The error, throughout, has been to call all three of this article’s central achievements, a swatch held out to an upholsterer, a serpent given a woman’s face, a reconstruction labelled dark matter, simply representation, as though one relation were doing the work in every case. Symbolic beings do not inhabit a world connected to their symbols by a single relation of correspondence. They inhabit several, partially independent relations of fit at once, and their symbolic worlds remain viable only for as long as cosmic fit holds good, mesocosmic and intersymbolic fit each remain internally coherent in their own right, and all three remain answerable to one another. The deepest problems, whether in art, in science, or in the ordinary business of furnishing a room, begin precisely at the point where they cannot.