Observed agreement does not determine a causal mechanism. SMM/DTT must derive stable dependencies among completed records and identify which interventions its source dynamics allows.
Three questions that causality can name
Why does a world exist? Why did one physical state change into another? What would happen if an experimenter altered one variable? These are questions about source dependence, dynamical mechanism, and intervention. They can be related without being interchangeable.
SMM principally motivates a common origin of physical multiplicity. A functional theory must also explain the ordinary dependencies by which its inhabitants act and infer. DTT adds the question of which generative dependencies survive completion as accessible causal relations. A statement that everything depends on one source does not by itself determine which laboratory manipulation changes which detector outcome.
A correlation with two different explanations
Let U be a fair binary variable. Model one assigns X=U and Y=U. Model two assigns X=U and Y=X. In passive observation both produce only (0,0) and (1,1), each with probability one half. Their observed joint distribution is identical.
Now intervene by setting X=0 while leaving U’s distribution unchanged. In model one, Y still equals U and is zero only half the time. In model two, Y becomes zero with certainty. The example demonstrates how a controlled intervention can distinguish a common cause from a directed influence when the intervention is correctly represented. The inference depends on the model of what the manipulation changes.
Earlier records are not automatically causes
A thermometer reading may change before an expanding rod is measured, yet both can respond to a common temperature change. The order of recorded detections also depends on travel times and instrument delays. Temporal precedence alone does not identify the mechanism.
An inner update ordering is similarly richer or different from the observer’s causal ordering. A source operation may precede a record without being separately controllable by the observer. The model should define its allowed interventions and their locations in the dynamics. This is especially important when a proposed hidden construction has an order that is absent from the effective relativistic description.
Repeated success and the domain of a law
If repeated preparations give a stable outcome frequency, that supports a probabilistic model within the tested conditions. It does not logically prove that the same rule holds under every possible preparation. The inference relies on assumptions about sampling, environmental stability, and relevant dependence between trials.
This leaves a positive strategy: estimate parameters in one set of conditions, freeze them, and calculate outcomes in others. A model earns explanatory reach when those additional results agree. A recreation account can permit unique complete events while retaining the same effective relation across a controlled class. Exact duplication of the entire universe is not required for reproducible science.
Make renewal carry the causal dependence
Write a source update as sn+1=F(sn,an), where a denotes an allowed intervention. Completed records are rn=R(sn). The model can calculate the consequences of replacing one action sequence by another, including effects delayed until later completion.
If the visible record omits predictive memory, the apparent transition law may depend on earlier records. Enlarging the retained state can restore a local description. SMM/DTT should explain this relation rather than declare all intermediate causes unreal. Effective causes can remain reliable descriptions of how interventions propagate through the shared source under a specified observation interface.
What a common-source explanation would add
One useful target is a relation among several causal responses that cannot be adjusted independently. If one parameter fixes the clock response, a transport delay, and a record’s relaxation rate, an experiment changing that parameter tests their joint dependence. Simply assigning each response its own temporal mechanism would provide less unification.
Read A69 with A82–A83 on future tests, A147 on projected memory, and A190–A194 on operational distinctions. Historical discussions such as al-Ghazali’s analysis of fire and cotton sharpen the difference between observed association and a claim of necessity. The modern intervention example here is a separate construction, not a mathematical result attributed to that historical text.
Continue with the articles
- Causality and Induction
Examine causality and induction as separate questions about explanation and inference. Relate them to a framework in which repeated records support laws inferred by observers.
- When two histories count as the same observation
Introduce completion and observational equivalence as distinct ways of identifying histories. Explain why the tests available to an observer determine which differences survive.
- Test languages and observable quotients
Explore how richer tests alter the distinctions retained by a record model. Introduce the draft’s correspondence between test languages and observational congruences.
- Eliminating hidden variables creates memory
Explain how removing the hidden sector produces an exact Volterra equation in the dual-time gauge model. A memory term can emerge even when the larger dynamics are specified locally in time.
- When memory can be represented by one time
Introduce finite Markovian embeddings of strictly proper real-rational memory. Explain why a hidden-memory description does not automatically imply an additional observable temporal degree of freedom.
- Empirical equivalence is not full structural equivalence
Separate equality of observed behavior from a global trivialization of the complete model. Explain why two exact questions require different evidence.
- How a protocol can reveal dual-time structure
Outline ways to move beyond the finite prepared linear interface: protocol-dependent access, noncommuting transport, or richer memory. Each route must be defined as an observable procedure.
- A checklist for claims of two physical times
Use the operational-reduction result to assess whether two time variables carry distinct empirical content. The key question is which observations cannot be reproduced by the reduced description.