Matthew Chenoweth Wright · Monolithic LLC

A research program in recursive physics.

Read the papers, inspect the equations and code, follow the formalization work, and help test what holds up. This site is a map into the Monolithic research corpus, with GitHub as its working record.

Recursive research loopA field of linked states proceeds through model, formalization, test, and revision.modelformalizetestrevise
Research cycle: propose a model, make its operations explicit, test it, and revise the claim against the result.
EFMWCore framework and equation archive
102Canonical Monolithic equations in the published set
Lean 4 + ZooFormalization and structured audit efforts

What is being studied

EFMW—Einstein–Feynman–Maxwell–Wright—is Matthew Wright’s proposed framework for modeling recursive coherence across physical and cognitive systems. The broader program turns that proposal into inspectable artifacts: equations, operational tests, formal proof obligations, simulations, and public research notes.

01 / Framework

EFMW and the field picture

Start with the central framework and its evolving equation set. Read each equation alongside its definitions, assumptions, units, and derivation status.

Open EFMW repository
02 / Canonical set

The Monolithic 102

A curated 102-equation corpus that organizes the program’s mathematical claims and their relationships.

Find the 102-equation materials
03 / Audit

The EFMW Zoo

A structured review method for checking claims against explicit operations, inputs, outputs, counterexamples, and anti-circularity requirements.

Open the post-156 audit
04 / Formal methods

Lean 4 formalization

Machine-checked statements can clarify what follows from encoded definitions. A Lean proof does not by itself establish that the definitions model nature.

Open Aristotle–102 Lean work
05 / Experiments

Control-degradation pilot

A narrow benchmark asks whether an EFMW-derived signal warns earlier than baseline monitors under matched operating conditions. Replication and independent review remain central.

Locate experiment code and records
06 / Research tools

Archimedes Engine

A research and reproducibility environment for organizing, running, and inspecting parts of the broader program.

Open Archimedes Engine

Research library

The GitHub organization is the canonical index for current papers, code, revision history, and supporting materials. The collections below offer a route through the work; repositories may evolve, so use each repository’s README and release history for its current contents.

Start with the source archive

Browse all public repositories to find the current equation index, Monolithic corpus, Lean projects, Zoo runs, and technical notes.

Connected lines of inquiry

These projects compare the EFMW approach with neighboring methods and explore its implications. Treat them as distinct research threads, not as established physical laws.

How to read the evidence

Different kinds of success answer different questions. This site keeps mathematical consistency, software behavior, empirical performance, and claims about the physical world separate so that collaborators can test the right thing.

Formal result

A statement proved from encoded definitions and assumptions. Inspect the code, theorem statement, and proof dependencies.

Computational result

A run or benchmark on stated data and settings. Reproduce it, compare baselines, and check sensitivity to seeds and metrics.

Physical claim

A proposed description of nature. It needs dimensional consistency, clear boundary conditions, discriminating predictions, and independent empirical tests.

Define

State symbols, units, domain, and assumptions.

Derive

Show each step and expose proof obligations.

Test

Freeze protocols, compare baselines, publish failures.

Replicate

Invite independent implementations and critique.

The material here represents a developing, author-led research program. A repository, theorem, simulation, or benchmark is not equivalent to peer review, independent replication, or accepted physical theory. Please assess each result from its linked methods and evidence.

Review, teach, build, fund

Monolithic LLC is seeking rigorous collaborators who can help make the program easier to inspect and more decisive to test. The most useful contribution begins with one bounded question and a result others can reproduce.

For researchers and developers

  • Choose one equation, proof obligation, or benchmark.
  • Record the exact repository revision and environment.
  • State the expected result before running the test.
  • Report positive, negative, and inconclusive outcomes.
  • Open a GitHub issue or discussion with the repository and artifact linked.

Choose a repository to review

For educators and students

Use the framework as a case study in modeling, formal methods, scientific inference, and reproducibility. Ask students to separate a theorem about a formal system from evidence about the world.

Suggested seminar sequence: read one equation note, inspect a Lean statement, reproduce a computational result, then write a short evidence assessment.

Fund a narrow, measurable next step

Monolithic welcomes research sponsorship, paid technical review, and pilot partnerships. A practical engagement should identify one deliverable—such as an independent Lean audit, a frozen benchmark replication, or a dimensional and boundary-condition review—with scope, budget, timeline, and publication terms agreed in writing. Funding supports the work; it does not determine the result.

No payment link or investment terms are represented here. To begin, contact Monolithic through the contact options on the GitHub profile or open a public issue in the relevant repository. Do not post confidential information in public issues.