A Systems Theory for Today

A living, forkable attempt to build a systems theory adequate to the present — by designed disagreement.

GOALS

Version: 0.2 · Status: Draft (author to ratify) · Last updated: Session 2

What we are trying to do, stated so plainly that we can later tell whether we did it.

The primary goal

Build a systems theory adequate to the world of the 2020s — a framework that can account, in a single coherent (if provisional and plural) order, for how today’s coupled systems produce today’s condition — using new methods, because the old method (one mind, one book) can no longer reach.

We treat this as a genuine open question, not a foregone conclusion. It is possible the honest finding is that no such theory can be built, or that only fragments can. That finding, rigorously reached, would itself be a result worth publishing.

The pressure-tests

A candidate theory earns credibility to the degree it illuminates these, without special pleading and without collapsing them into a single glib cause — and, since Session 2, to the degree it draws the arrows between them: how a driver produces a dynamic produces a symptom, and how symptoms feed back.

The founding seven (honored; the project’s origin): fertility, the dopamine/attention economy, the insufficiency of “post-truth,” the meta-question of why now, climate, coordination failure, and the resurgence of populism/authoritarianism — beneath them a master puzzle: the loss of shared belief and coherent understanding.

The expanded set (Session 2, panel/SESSION_2_PRESSURE_TESTS.md). The panel re-debated the seven, admitted five more, and — over live dissent — restructured the flat list into four coupled layers plus a master condition, presented as a contested causal hypothesis, not a taxonomy. All seven are preserved; their new homes are noted. Additions marked [+].

  • M — The coherence vacuum (master condition; whether it is the base or the summit is left open — Q-002). The loss of shared meaning, belief, and a common frame for understanding. (Absorbs old #4, “why now,” as the same demand from the inside.)

  • Layer 1 — Drivers (deep, slow, generative)

    • D1 Acceleration — change outrunning adaptation (the adaptation gap; reframes old #4).
    • D2 Optimization dynamics — self-reproducing optimizers capturing human drives.
    • D3 Wealth pump / inequality [+] — structural wealth concentration and elite overproduction.
    • D4 Machine intelligence [+] — computation reshaping labor, cognition, epistemics, coordination.

  • Layer 2 — Dynamics (couplings the drivers produce)

    • Y1 Epistemic breakdown (old #3) · Y2 Coordination failure (old #6) · Y3 Institutional & legitimacy decay [+] · Y4 Ecological overshoot (old #5, generalized beyond climate).

  • Layer 3 — Symptoms (what we feel)

    • S1 Fertility collapse (old #1) · S2 Attention/dopamine economy (old #2) · S3 Political form: populism (old #7) · S4 Anomie / the loneliness epidemic [+] — the felt face of M.

  • A cross-cutting reflexive test (added Session 4k, at the author’s ratification)

    • R — Register/care fairness [+]. Does the account — and the apparatus producing it — systematically under-weight care and relational registers (the reproductive, the domestic, the relational, the register in which a life is actually lived and lost) relative to analytic and systemic ones? This is promoted from the standing D-005 dissent (Le Guin, “whose crises are on the list”; the care debt named three cycles and left unpaid) to a falsifiable test: it is falsified if a register audit of the corpus finds care/relational conditions carried at parity with analytic ones; it confirms register-blindness if they are systematically under-represented, or spoken-about but never from. It sits across the M/D/Y/S layers, not inside one, because it tests the list itself — and its own confounder is stated plainly: the apparatus scoring it is the same one under audit (L-015), so like every internal instrument it inherits the one-aquifer limit and is not inferential until a foreign vantage (the S4k concession). The fourteenth pressure-test.

Tracked as effects/amplifiers, not top-level tests (revisit): migration, financialization (folded into D3), pandemics/biosecurity, geopolitical/war risk, energy transition (folded into D4/Y4).

The layering is itself contested — Marx rejects the driver/symptom cut, Nietzsche puts M at the summit not the base, Le Guin notes the ordering encodes a worldview (all preserved in SESSION_2_PRESSURE_TESTS.md and logs/OPEN_QUESTIONS.md). The visual form is the concept map in docs/DIAGRAMS.md §4. The signed causal hypothesis over these thirteen layered tests — the arrows drawn with direction, sign, and (where known) delay, deliberately falsifiable and still contested — is docs/PRESSURE_TESTS_CAUSAL_HYPOTHESIS.md (ratified as the working hypothesis, Session 4e). (The fourteenth test, R — register/care fairness, added S4k, is deliberately cross-cutting and reflexive — it audits the list and the apparatus, not a node in the signed loop — so it does not appear as an arrow in the causal graph.)

Secondary goals

  • Answer the historical question honestly (done in draft: HISTORY_OF_SYSTEMS_THEORIES.md): what total systems existed, who built them, when they fell, and what is being tried now.
  • Situate ourselves in the living field honestly (done in draft: LANDSCAPE_OF_CONTEMPORARY_SYSTEMS_THEORIES.md): survey the serious contemporary attempts — cliodynamics, complexity science, Luhmann, polycrisis/metacrisis, Big History, metamodernism, integral theory, and more — say what each reaches and can’t, and locate the precise gap we aim at. Public-facing (site + reports), because a theory that ignores its neighbors is not adequate to today.
  • Assemble and operate a panel of real thinkers as designed disagreement (panel/).
  • Produce our own candidate theories — at least two, ideally three — as living seeds others can extend or break (outputs/CANDIDATE_THEORIES.md).
  • Define our terms — “understanding,” “today,” and “systems theory” are contested and must be argued, not assumed (docs/GLOSSARY.md).
  • Build toward code — migrate the candidate theories from prose to machine-readable models to runnable simulations (ARCHITECTURE.md §6).
  • Make it public and forkable — GitHub repository + website, so others contribute, contradict, and extend (CONTRIBUTING.md).
  • Make it learn — a catches-and-learnings loop so the project corrects itself over time (logs/).

What success looks like (and how we’ll know)

Because “we built a total theory of everything” is not a checkable claim, we define success operationally:

  • S1 — Illumination. For each pressure-test (now fourteen — thirteen in four layers, plus the cross-cutting register/care test R), the project offers a systems-level account — and draws the couplings between them — that a thoughtful skeptic finds more explanatory than the loose talk currently on offer (measured against the best of the field in docs/LANDSCAPE_OF_CONTEMPORARY_SYSTEMS_THEORIES.md, not its memes). Measured by red-team review, not self-assessment.
  • S2 — Falsifiability. At least one candidate theory generates a claim that could, in principle, be shown wrong — a prediction, a mechanism, or a model output. (Turchin’s standard: a theory that risks nothing explains nothing.)
  • S3 — Coherence with honesty. The candidate theories hang together and the project openly records where they don’t, where they were overturned, and what remains unexplained.
  • S4 — Livability. The repository is genuinely usable by a stranger: a contributor can find where to add, fork, and propose a rival without breaking the structure.
  • S5 — Reflexivity that pays off. The method itself (designed disagreement + synthesis + continuous revision) demonstrably catches errors an individual author would have missed — logged in logs/.

Non-goals (to keep us honest)

  • Not a claim to finality. Every output is provisional; “proto-synthesis,” in the metamodern phrase.
  • Not neutrality-as-emptiness. The panel takes sides; the project has views. (See GROUND_RULES.md.)
  • Not prophecy. We explain and model; we do not predict a fixed future. Where we forecast, we forecast conditionally and to be acted upon.
  • Not a single doctrine imposed by the chair. The chair synthesizes and does not vote; the author ratifies; contributors dissent.

The reflexive goal (the one that ties form to content)

Demonstrate, by building it, that a systems theory for today can be held as a living commons rather than authored as a monograph — that the way past the impossibility of the old method is a procedural totality: many minds, in structured disagreement, continuously integrated. If the project works, it is itself a small instance of its own third candidate theory.