reality: what and how
Reality: What, How, and Under What Conditions
A demonstrative epistemic essay on context-sensitive representation
Status
Living demonstrative essay within the Dot Theory research environment
Original title
Reality: What and How — Understanding Reality: It Is Not What We Thought
Originally published
23 March 2026
Revised
25 July 2026
Current claim status
Programmatic, representational and empirically testable
Canonical status
Non-canonical explanatory material
Revision and provenance note
This page preserves the originating question and principal state–context distinction of the earlier essay while revising its present framing in light of the subsequent development of Constitutional Physics.
The revision:
distinguishes demonstration from empirical validation;
separates representational claims from physical and ontological claims;
narrows propositions previously stated too generally;
makes the proposed empirical comparison explicit;
and identifies what a positive, negative or null result would establish.
It does not retrospectively claim that the later constitutional architecture was fully formalised when the original page was written.
Where this living explanatory page differs from the fixed published corpus, the relevant versioned publication provides the controlling documentary statement.
Website orientation: https://www.dottheory.co.uk/paper/website-orientation
Physics Programme: https://www.dottheory.co.uk/physics-programme
Research Programme: https://www.dottheory.co.uk/paper/research-programme
Introduction
Sharp titles create expectations of strong arguments.
This page does not present a total theory of reality. It presents a bounded representational question and demonstrates how Dot Theory approaches that question.
Under what conditions is a representation sufficiently complete for the scientific, computational or communicative task assigned to it?
The originating intuition is that scientific models may remain extraordinarily useful while omitting contextual conditions that affect how their represented states are constituted, interpreted or evolved.
That possibility does not imply that science is generally mistaken. It does not imply that context is absent from existing scientific practice. It does not imply that every model should contain every available variable.
It asks something narrower:
Are there circumstances in which outcome-relevant contextual structure is used implicitly, treated as external or discarded from the explicit unit of representation, and would representing that structure improve the model under declared conditions?
This is a question about representational adequacy.
It becomes a physical claim only when applied to a specific physical system with an appropriate derivation, mechanism and empirical discriminator. It becomes a computational claim only when the proposed representation is implemented and comparatively evaluated. It becomes an ontological claim only if further argument and evidence justify that extension.
This page demonstrates the movement among those levels without treating them as interchangeable.
Applied comparison: https://www.dottheory.co.uk/paper/context-sensitive-modelling-in-practice
The short claim
Scientific models can be useful, predictive and technically successful while remaining incomplete relative to a particular question.
Practical success and representational completeness are not the same property.
A map may preserve coordinates and distances extremely well while omitting terrain, traffic, accessibility, social meaning or the conditions under which a journey will actually occur. That does not make the map wrong. It makes its adequacy relative to purpose.
A scientific model may remain practically successful while being structurally incomplete where outcome-relevant contextual conditions are used implicitly but omitted from the explicit representation on which prediction or interpretation depends.
This is not a claim that all scientific models represent only state. Many established scientific disciplines already incorporate environmental conditions, boundary conditions, measurement settings, regimes, covariates and observer-dependent information.
The question is whether these are always represented with sufficient precision at the level where meaning, comparison and further computation occur.
Dot Theory therefore does not propose replacing successful scientific methods. It asks whether those methods may sometimes be operating upon a representational object whose governing context is only partly visible.
A good method may be applied to an incompletely specified object. The method need not be discarded. The object may need to be represented more fully.
From state to state under structure
Many physical and computational models begin with a state description.
Let:
ψ
represent the declared state of a system.
Let:
μ
represent contextual structure relevant to the declared task. Depending upon the application, this may include:
environmental or boundary conditions;
measurement configuration;
regime or phase;
operator position;
informational accessibility;
data-selection conditions;
model permissions;
interpretative assumptions;
or other declared variables affecting the relationship between state and outcome.
An extended representation may then be written as:
Ψ = (ψ, μ)
The ordinary state description is recovered through a projection:
π(Ψ) = ψ
Under what circumstances does the projection from Ψ to ψ discard information required for the task being performed?
This does not establish that every omitted contextual variable matters.
Some contextual variables may be irrelevant.
Some may duplicate information already encoded in the state.
Some may improve explanation but not prediction.
Some may increase complexity without producing sufficient benefit.
Some may not be measurable or communicable with adequate reliability.
The hypothesis is conditional:
Where contextual structure materially governs observable outcomes, an explicit context-sensitive representation may outperform a reduced representation that omits or obscures that structure.
The distinction is not simply between an object and no object. It is between an object represented through selected state and that object represented under declared structure.
Reality is represented as an object under structure.
That is a claim about the constitutional form of communicable representation. It should not, by itself, be treated as a completed claim about reality in itself.
On Reality: https://www.dottheory.co.uk/paper/on-reality
On Boundaries: https://www.dottheory.co.uk/paper/on-boundaries
The uncomfortable consequence
The proposed consequence is not that science is wrong. It is that successful models may sometimes be incomplete in a specific and recoverable way.
A model may approximate Ψ = (ψ, μ) while representing only ψ explicitly.
The effects of μ may already be present through calibration, experimental procedure, parameter selection, boundary conditions, institutional knowledge or tacit interpretative practice. The model may therefore work because its wider scientific environment supplies structure that the explicit computational object does not itself preserve.
As instruments improve, datasets become richer and interactions among frameworks become more complex, previously implicit contextual structure may become visible and representable.
When that occurs, the relevant question is not whether context exists in some general sense. It is whether making a particular context explicit produces a demonstrable improvement.
That improvement might concern:
predictive accuracy;
calibration;
explanatory coherence;
transfer between environments;
interpretative stability;
recovery by independent operators;
reduction of hidden assumptions;
or prevention of unsupported extension.
The claim must remain local to the task and metric declared.
A model cannot be described as more complete merely because it contains more information. Additional context has costs.
increased data requirements;
greater computational load;
parameter growth;
storage and bandwidth;
interpretative burden;
privacy risk;
documentary complexity;
and reduced generalisability.
A useful context-sensitive model must therefore justify the additional structure it introduces.
The empirical question
A minimal context-dependence condition can be expressed as:
∃ μ₁, μ₂ such that P(O | ψ, μ₁) ≠ P(O | ψ, μ₂)
There exist contextual conditions under which apparently equivalent declared states are associated with different outcome distributions.
This result alone would establish only that the selected context matters under the tested conditions. It would not establish Dot Theory as a complete theory, prove a new physical ontology, or show that every state-centred model is inadequate.
The stronger empirical comparison requires two model classes:
Reduced model
P(O | ψ)
Context-sensitive model
P(O | ψ, μ)
The two models should be evaluated through:
matched source data;
equivalent training and testing conditions;
preregistered outcome measures;
independent or out-of-sample evaluation;
declared model-complexity controls;
transparent treatment of missing information;
and accounting for the cost of acquiring and representing μ.
Does explicit representation of the declared contextual structure produce a reproducible and worthwhile improvement over the reduced model after the additional informational and computational burden has been taken into account?
A positive result would support the bounded proposition tested. A partial result might show improvement only for particular outcomes, contexts or participant classes. A null result would indicate that the selected contextual representation did not materially improve the declared task under the tested conditions. A negative result might show that the added structure reduced performance, increased instability or imposed costs greater than its benefit.
Each is scientifically informative. None should be repaired after the result by changing the meaning of context, the success criterion or the proposition being tested.
Why this is not a total theory of reality
A complete and self-contained theory of reality faces a structural difficulty. It must either exclude some of the contextual conditions governing its own construction and validity, or attempt to include those conditions within itself.
If it excludes them, its claimed completeness is qualified. If it includes them, new conditions governing that inclusion may arise.
This page does not offer that observation as a proof that a complete theory is impossible in every conceivable sense. It uses it to motivate a more pragmatic objective:
Not possession of a final description of reality, but construction of representations whose conditions, transitions and limits are sufficiently visible for their declared purposes.
The programme therefore does not begin by asking only “What is reality?” It also asks:
Through which operators, conditions and accessible structures does a claim about reality become communicable, comparable and available for further use?
That shift does not make reality merely subjective. It makes the conditions of its representation part of the scientific object under examination.
Dot Theory as an operational theory of communicable reality
In scientific and communicative practice, human beings do not work with reality without mediation. They work through observations, measurements, instruments, models, mathematical structures, language, simulations, memories, institutional records and other accessible representations.
Each is produced through an operator acting upon an accessible domain under particular conditions.
Disagreements may therefore arise from more than incompatible conclusions about reality itself. They may arise from differences in:
informational accessibility;
operator position;
assumptions;
definitions;
projections;
comparison interfaces;
admissibility rules;
or the level at which a claim is being made.
Dot Theory shifts attention from the pursuit of context-free certainty towards the governance of communicability.
Under what conditions does a claim become communicable, comparable and admissibly related to other claims?
In this bounded sense, Dot Theory may be understood as an Operational Theory of Communicable Reality.
It is not primarily a theory declaring what reality finally is. It is a framework for examining how represented reality becomes available for lawful interaction among people, disciplines, institutions and computational systems.
The Dot Operator
The Dot Operator is the contextual-admissibility operation through which Dot Theory examines a represented object or claim. It presently performs seven related functions.
1. Localise
Every claim, observation, framework, model or belief must first be situated.
The operation asks where it came from, under what conditions it was generated, which operator or process produced it, which domain it claims to describe, and at what representational level it operates.
Before comparison becomes meaningful, locality must be established.
2. Expose
Frameworks often contain commitments that have become invisible through familiarity.
Assumptions may be treated as observations. Interpretations may be treated as facts. Procedures may appear neutral after their governing choices have been forgotten.
The Dot Operator returns these conditions to examination.
3. Compare
Once locality and governing commitments have been declared, comparison becomes possible.
Comparison does not require agreement. It requires sufficiently explicit conditions.
The objective is not to reduce every framework to one vocabulary. It is to establish what is genuinely shared, what differs and which differences prevent direct comparison.
4. Bridge
Independent frameworks may preserve different aspects of the same observation or problem.
Where a relationship preserves the meaning required by both sides, a bridge may be proposed. Where preservation fails, the failure should be recorded rather than concealed.
A failed bridge may reveal an important boundary.
5. Preserve residuals
Every projection preserves some structure and leaves other structure behind.
Dot Theory treats these residuals as legitimate objects of analysis rather than automatically classifying them as noise or error.
A residual may identify inaccessible information, representation loss, framework incompatibility, unmodelled context or a possible direction for further enquiry. It may also prove irrelevant. Its status must be evaluated rather than presumed.
6. Evaluate admissibility
Not every proposed relationship is legitimate. Not every analogy preserves meaning. Not every translation is reversible. Not every combination of frameworks is computationally or scientifically warranted.
A claim becomes admissible not because it is certain, but because the conditions under which it is made and the consequences it is permitted to support have been sufficiently declared.
7. Govern interoperability
Knowledge increasingly emerges through interactions among disciplines, frameworks, institutions, datasets and human or artificial operators.
The purpose of interoperability governance is not to collapse their distinctions. It is to permit productive interaction while preserving differences that remain scientifically, computationally or constitutionally important.
Operational-Admissibility Matrix: https://www.dottheory.co.uk/paper/admissibility-matrix
Lexicon: https://www.dottheory.co.uk/paper/lexicon
What the page demonstrates
This page performs the procedure it describes.
It begins with an intuitive proposition: Reality is not fully represented by state alone.
It then localises that proposition as a question about representational adequacy.
It exposes the distinction between state and context.
It formalises that distinction as Ψ = (ψ, μ).
It identifies a projection through which contextual structure may be omitted: π(Ψ) = ψ.
It states a candidate empirical condition: P(O | ψ, μ₁) ≠ P(O | ψ, μ₂).
It then restricts the conclusion to what that comparison could establish.
intuition → localisation → distinction → formalisation → comparison → test → bounded interpretation
That movement is itself an example of the Dot-theoretical method.
The demonstration may succeed even if the substantive hypothesis later fails. A failed hypothesis can still have been properly constituted, clearly tested and correctly bounded. That is part of the scientific value of the procedure.
What the page does not demonstrate
This page does not establish that:
reality is ontologically composed of ψ and μ;
all existing scientific models are state-only;
contextual representation will improve every model;
every disagreement is caused by different operators or contexts;
a more complex representation is automatically more accurate;
context can always be measured reliably;
Dot Theory has been empirically validated;
or a complete physical theory has been disproved.
Those propositions would require arguments and evidence beyond this page.
The distinction between demonstration and validation is therefore essential.
This page demonstrates the representational distinction, the proposed transition, the operation of the method, and the form of evidence required for comparative evaluation. It does not predetermine the result.
A disciplined form of contextualisation
Most systems of reasoning are ultimately asked: “What conclusion should I accept?” Dot Theory asks an earlier question:
Under what conditions may this conclusion be made, communicated, compared or combined without losing the meaning required by the task?
This can be understood as a disciplined form of contextualisation: not wisdom as authority, possession of final truth or accumulated information alone, but wisdom as the continual practice of:
localising;
exposing;
comparing;
bridging;
preserving residuals;
evaluating admissibility;
and governing interoperability.
The objective is not possession of reality. The objective is the continual refinement of communicable representations of reality.
Reality may exceed every operator presently available to us. Communicable reality is the domain in which available representations become capable of disciplined interaction. That domain includes science, mathematics, language, law, governance, education, computation and ordinary human understanding.
Where this leads
The strongest recoverable conclusion is not:
Reality is no longer an object.
It is:
Every communicable and computable representation presents an object under conditions.
Or, more intuitively:
No represented thing appears without some structure governing how it becomes available.
Borrowing loosely from Kant, the relevant scientific object is not simply Ding an sich, but Ding unter Struktur: the thing as made available under declared structure.
When governing conditions remain inaccessible, they may appear fixed, irrelevant or indistinguishable from the object itself. When they become accessible and measurable, they may become representable. When they become representable, they may become computable. When they become computable, their contribution can be tested.
Accessibility therefore helps determine what can be represented. Representation helps determine what can be computed. The completeness required is always completeness relative to a declared operation, question and evidential standard.
Closing
Science has achieved extraordinary explanatory, predictive and practical success by constructing stable objects of enquiry and applying disciplined methods to them.
This work does not oppose that achievement. It asks whether some contemporary problems require the contextual conditions of those objects to be represented more explicitly.
Where outcomes depend upon both state and context, explicitly representing the relevant context may improve prediction, interpretation or communication.
This page does not establish that context-explicit representation will outperform every existing scientific model. It demonstrates the distinction, identifies the representational transition and states the form of evidence required to evaluate it.
The next question is empirical.
Does the additional structure produce a reproducible benefit under declared conditions, and is that benefit sufficient to justify the additional informational, computational and documentary load?
That can be formalised. It can be tested. It can fail. And where it produces demonstrable benefit, it can be incorporated into scientific and computational practice without requiring the premature declaration of a total theory of reality.
That is non-trivial. It is, in my view, worth investigating.
Thank you for your time and consideration.
—Stefaan Vossen⊙
Revision record
Current revision
Date: 25 July 2026
Revision class: Substantive explanatory recontextualisation
Canonical effect: None
Empirical effect: None
Historical effect: Original developmental state preserved through this notice
Material changes
replaced the implication that science generally models reality incorrectly with a bounded representational-incompleteness hypothesis;
distinguished a claim about communicable representation from a claim about reality in itself;
clarified that many scientific frameworks already incorporate contextual variables;
defined ψ, μ, Ψ and the projection π more precisely;
separated evidence of context dependence from evidence of comparative modelling benefit;
added preregistration, out-of-sample testing, complexity control and load accounting as conditions of a meaningful comparison;
corrected the Dot Operator from six to seven listed functions;
distinguished the page’s demonstrative success from empirical validation of its substantive hypothesis;
identified positive, partial, null and negative outcomes as admissible;
preserved the original programme direction while narrowing unsupported physical and ontological extensions.
Earlier version
The earlier page should be preserved by one of the following means:
an archived webpage;
a dated PDF or text copy;
a repository deposit;
or a locally retained file identified by filename, date and cryptographic hash.
The earlier version should not remain the controlling explanatory page, but it should remain recoverable as part of the programme’s development history.
