New Scientific Method
8 Epistemic Layers
Forward + Reverse
5 Death Conditions
7Q Method Explorer
Seven questions forward: classify anything. Seven questions reversed: prove anything. Q0: arrive humble or don't arrive at all.
Q0 — Presupposition
Q0 — PRECONDITION
The Inquirer Cannot Be the Ground of Inquiry
Arrive open or don't arrive at all.
Before the first question, there is a posture. You do not arrive decided. You arrive open to whatever the evidence produces, including results that disconfirm your priors. This is not humility as virtue — it is an operational requirement. Every dead framework committed the exact error Q0 prevents: positivism arrived decided that metaphysics was meaningless; behaviorism arrived decided that consciousness was irrelevant. Q0 is the silence before the first letter. The empty soil before any seed lands.
Q0 ∉ {Q1..Q7} — sits outside the sequence
posture(decided) → dead framework
posture(open) → breakthrough potential
posture(decided) → dead framework
posture(open) → breakthrough potential
Q0 — OPERATIONAL
What Q0 Prevents
Historical test: every framework that ossified arrived decided.
Einstein didn't set out to disprove Newton — he resolved a genuine tension. Planck didn't want quanta — the data forced him. Q0 is the weakest operational element (non-enforceable) and the most important structural one. Cannot be scored, only acknowledged.
Q1 — Definition
Q1.1 — CLASSIFICATION
Name It. Type It. Tier It.
Undefined → defined. Create the object before testing it.
Q1 forces the inquirer to name the claim, classify its type, and assign it a tier. Misclassification at Q1 propagates through every subsequent question. Has it been named before — and does that name carry assumptions you haven't verified?
Types: causal · descriptive · ontological · mathematical · predictive · mechanistic · normative
Q1.2 — TIER
Where in the Knowledge Hierarchy?
Tier determines how a claim is tested.
A foundational claim (axiom, postulate, definition) needs different kill conditions than a derived claim (theorem, corollary, hypothesis). A hypothesis treated as an axiom will never be traced to its dependencies — because axioms don't have dependencies, they ARE dependencies.
Foundational: axiom · postulate · definition
Derived: theorem · corollary · hypothesis
Constraint: boundary condition
Support: evidence bundle · Connection: bridge · relationship
Derived: theorem · corollary · hypothesis
Constraint: boundary condition
Support: evidence bundle · Connection: bridge · relationship
Q1 — GR EXAMPLE
GR Is Causal, Not Descriptive
Q1 reclassified GR from descriptive to causal — a stronger commitment.
Standard textbook presentation treats GR as a description of spacetime geometry. Q1 reclassifies it: GR asserts that mass-energy CAUSES curvature. That is a causal claim — a stronger commitment than "spacetime is curved." The reclassification immediately sharpens the kill conditions.
classify(GR) = causal > descriptive
Causal = removing the cause should remove the effect
This is testable in ways "descriptive" is not.
Causal = removing the cause should remove the effect
This is testable in ways "descriptive" is not.
Q2 — Definition
Q2.1 — DOMAIN ANCHOR
Anchor in Domain and Scale
Defined → located. Place the claim in conceptual space.
Q2 establishes primary domain, additional domains, and scale of operation. The critical question: does this claim appear in more than one domain? Cross-domain presence triggers the isomorphism test.
Domains: physics · theology · consciousness · info theory · math · biology · economics
Scale: quantum → molecular → neural → individual → social → civilizational → cosmic
Scale: quantum → molecular → neural → individual → social → civilizational → cosmic
Q2.2 — ISOMORPHISM TEST
Analogy or Structural Identity?
If the math preserves topology, boundary conditions, and conservation laws under variable substitution → isomorphism, not analogy.
Analogies suggest similarity. Isomorphisms demonstrate structural identity. If the same structural relationships hold when you swap domain labels — the same tensor equations, the same conservation laws — that's not metaphor. That's structure. Riemannian geometry appears in physics (GR), ML (manifold learning), economics (curved utility), and information theory (Fisher geometry). Same mathematics in all four domains.
iso(A,B) iff ∀ relations R: R_A ↔ R_B under variable substitution
Analogy: "gravity is like love" — metaphorical, non-constraining
Isomorphism: "the same tensor equations describe both" — structural, prediction-constraining
Analogy: "gravity is like love" — metaphorical, non-constraining
Isomorphism: "the same tensor equations describe both" — structural, prediction-constraining
Q3 — Definition
Q3.1 — PRECISION
One Sentence. No Hedging.
Located → stated. Every claim has a single falsifiable core.
Q3 forces the claim into a single precise statement. Classify by certainty level (proven, well-supported, tentative, speculative) and scope (universal, domain-specific, context-dependent). The opponent restatement test: can a serious, informed opponent restate your claim accurately? If not, it's not precise enough to test.
Q3.2 — NEGATION TEST
Every Claim Carries Its Own Death Warrant
Auto-generate the logical opposite. Hidden bundled claims surface here.
Q3 automatically generates the negation: ¬(A ∧ B) = ¬A ∨ ¬B. If the negation decomposes the original into independently testable components, you've found bundled claims hidden inside a single statement. GR's negation: "Mass-energy does NOT curve spacetime, OR objects do NOT follow geodesics." Two independently killable components — found automatically.
¬(A ∧ B) = ¬A ∨ ¬B — negation splits bundled claims
GR negation: ¬(curvature causation) ∨ ¬(geodesic following)
GR negation: ¬(curvature causation) ∨ ¬(geodesic following)
Q4 — Definition
Q4 — CHANNEL 1: PS
Phenomenon Strength — What Do We Reliably Observe?
Reproducibility × 0.4 + Effect size × 0.3 + Measurement quality × 0.3
The rain that falls on the tree. This is what science already does well: measurement, replication, signal detection. Reproducibility is weighted highest because if it can't be repeated, it doesn't exist as evidence.
PS = reproducibility×0.4 + effect_size×0.3 + measurement_quality×0.3
Q4 — CHANNEL 2: ED (THE WHY-PENALTY)
Explanatory Depth — Do We Know WHY?
Mechanism clarity × 0.4 + Constraint consistency × 0.3 + Scope × 0.3. ED = 0 caps evidence at 50%.
The sunlight that tells the tree which direction to grow. This is what science systematically avoids. Every dead framework refused this question. Every breakthrough came from someone who held it. The why-penalty: if you observe without explaining, you cannot have full credit. Pre-1927 scientists who asked WHY moved the needle. Post-1927 instrumentalism correlates with fragmentation.
ED = mechanism_clarity×0.4 + constraint_consistency×0.3 + scope×0.3
ED = 0 → evidence capped at 50% regardless of PS
ED = 0 → evidence capped at 50% regardless of PS
Q4 — CHANNEL 3: EC
Experiential Coherence — Does It Produce Stable Structure?
Internal consistency × .25 + Longitudinal stability × .25 + Behavioral transformation × .25 + Intersubjective pattern × .25
The soil moisture the roots actually draw from. Not "I feel strongly" — does the evidence produce consistent downstream effects across time, behavior, and agents? Does the insight change how people act? Do multiple observers report the same structure across different contexts?
EC = consistency×.25 + stability×.25 + transformation×.25 + pattern×.25
Q4 — THE COMPLETENESS FACTOR
E_final = PS × CF — The Three Channels Multiply
CF = (0.5 + 0.5 × ED) × (0.5 + 0.5 × EC). A tree with infinite rain but no sunlight dies.
The three channels multiply, not add. You cannot compensate for zero in one channel by maxing another. ED = 0, EC = 0 → CF = 0.25 → evidence capped at 25%. ED = 1, EC = 1 → CF = 1.0 → full credit earned. Observed ≠ Explained. Repeatable ≠ Understood.
CF = (0.5 + 0.5×ED) × (0.5 + 0.5×EC)
E_final = PS × CF
ED=0, EC=0 → CF=0.25 (25% ceiling)
ED=1, EC=1 → CF=1.0 (full credit)
E_final = PS × CF
ED=0, EC=0 → CF=0.25 (25% ceiling)
ED=1, EC=1 → CF=1.0 (full credit)
GR Evidence — Kill Condition Results
| Evidence | Result | Formal |
|---|---|---|
| Mercury perihelion precession | SURVIVED | Δφ = 43.0″ ± 0.1″ per century |
| Gravitational lensing (Eddington 1919) | SURVIVED | θ = 4GM/(c²b) |
| Gravitational waves (LIGO 2015) | SURVIVED | Chirp waveform matched |
| GPS time dilation | SURVIVED (daily) | +38 μs/day net correction |
Q5 — Lemma
Q5.1 — CHAIN TERMINUS
Where Does the Chain End?
Supported → load-tested. Every dependency chain has a terminus.
Q5 traces the dependency chain all the way down. Terminus types: Axiom (self-evident) → solid ground. Brute fact (accepted without proof) → honest but ungrounded. Circularity (loops back) → STRUCTURAL VULNERABILITY FLAGGED. Open (never terminates) → INCOMPLETE FLAGGED.
terminus(axiom) = solid ground
terminus(brute_fact) = honest
terminus(circular) = FLAGGED — structural vulnerability
terminus(open) = FLAGGED — incomplete
terminus(brute_fact) = honest
terminus(circular) = FLAGGED — structural vulnerability
terminus(open) = FLAGGED — incomplete
Q5.2 — THE FRAGILITY MAP
If This Foundation Fails, What Happens?
Structural engineering load test applied to ideas.
For each dependency: immediate collapse (critical) · graceful degradation (the claim still partially works) · independent survival (the dependency wasn't actually needed). GR example: Q5 surfaced GR's hidden dependency — the smooth manifold assumption. At Planck scale (~10−³&sup5; m), spacetime might be discrete. This has been driving unsolved physics for a century. Q5 surfaced it in one question.
GR depends on: equivalence principle · Riemannian geometry · c=const · E-mom conservation · smooth manifold
Hidden: smooth manifold assumption (unverified at Planck scale)
→ root of the quantum gravity problem
Hidden: smooth manifold assumption (unverified at Planck scale)
→ root of the quantum gravity problem
Q6 — Theorem
Q6.1 — PREDICTIONS
What Must Be True If This Claim Is True?
Grounded → generative. Confirmed/untested prediction ratio = generative maturity.
A claim that forces nothing downstream explains nothing — it occupies conceptual space without generating knowledge. Q6 separates confirmed predictions from untested ones. GR forced black holes (Schwarzschild 1916), gravitational waves (detected 2015), GPS corrections (daily) — all consequences the math generated that Einstein himself didn't plant.
generative_maturity = confirmed_predictions / total_predictions
Q6(claim) = ∅ → explanatory failure signal
Q6(claim) = ∅ → explanatory failure signal
Q6.2 — CROSS-DOMAIN FORCING
Where Do Consequences Land in Other Fields?
If a Q6 consequence preserves structural relationships in another domain → isomorphism candidate. Loops back to Q2.
This is the discovery mechanism for cross-domain connections. Standard disciplinary boundaries hide these connections. The framework makes them structurally visible. When a consequence appears in another domain and preserves the same mathematical structure under variable substitution, an isomorphism candidate has been detected — loop back to Q2 for verification.
Q6 consequence ∈ domain_B → flag for Q2 isomorphism test
Q2 ↔ Q6 form a detection loop
Q2 ↔ Q6 form a detection loop
Q7 — Axiom
Q7 — DEATH CONDITION #1
Self-Refutation
P → ¬P. The fastest death condition to check.
The claim destroys itself when stated. The denial presupposes what it denies. "There is no truth" is itself a truth claim. "Logic is unreliable" requires logic to state. If the claim self-refutes, nothing else matters. Many philosophical positions fail here silently, never examined because the framework never demanded it.
Q7 — DEATH CONDITION #2
Infinite Regress
answer(Q) → new instance of Q → ∞
The claim pushes the question backward without ever resolving it. Each answer generates a new instance of the same question. Not the same as a deep dependency chain — regress means the chain never terminates by design. Some frameworks embrace this; the 7Q system flags it as a structural failure.
Q7 — DEATH CONDITION #3
Empirical Contradiction
prediction(claim) ≠ observation → falsified
The claim contradicts observed data. This is the traditional mode of falsification — the one Popper formalized. Necessary but not sufficient alone: a claim can survive empirical testing and still fail on other conditions. The 7Q system adds four more death conditions beyond Popper's one.
Q7 — DEATH CONDITION #4
Logical Incoherence
claim ∧ ¬(consequence of claim) → incoherent
The claim contradicts something already established in its own chain — a prior dependency, a forced consequence, or itself at a different point in the argument. Different from self-refutation (which is immediate). This catches delayed contradictions that only appear when the full chain is traced.
Q7 — DEATH CONDITION #5
Explanatory Failure
Q6(claim) = ∅ — the claim stops generating consequences
The claim cannot answer forced downstream questions from Q6. It stops generating. It runs out of road. Kuhn described this as anomaly accumulation — the 7Q system formalizes it. A theory that stops generating is dying. Different from empirical failure — the data might still fit, but the theory can no longer explain.
Q7 — ADVERSARIAL TEST
Has Anyone Actually Tried to Kill It?
Surviving sympathetic review ≠ surviving hostile review. Adversarial status: tracked.
A claim that survived adversarial testing by informed opponents is categorically different from one examined only by sympathetic parties. Most claims in most fields have NOT been adversarially tested. The 7Q system requires this field to be declared. A claim with no adversarial history is unproven, not confirmed.
T — Theorem
SYNTHESIS — TRUTH SCORE
T = w&sub1;·S + w&sub2;·E + w&sub3;·L + w&sub4;·D + w&sub5;·P + w&sub6;·C
Six variables, all derived from 7Q fields. No external judgment needed.
S ← Q7 survivability · E ← Q4 E_final with why-penalty · L ← Q3 precision × Q4 type · D ← Q5 dependency stability · P ← Q6 confirmed prediction ratio · C ← Q2 cross-domain coherence. Every variable comes from fields the inquirer already filled. No external judgment needed — the math handles it.
T = Σ w_i · V_i where V ∈ {S, E, L, D, P, C}
Maps to: T=χ, confidence=κ, cluster variance=ρ
Maps to: T=χ, confidence=κ, cluster variance=ρ
SYNTHESIS — MODULATOR
T_enhanced = T × (0.6 + 0.4 × 7Q_score)
Zero 7Q markers = 60% of truth score. Full 7Q = 100%.
The 7Q framework score itself modulates the truth score. This means: using the method makes your work score higher. Not because it's biased — because structured epistemic practice IS better epistemic practice.
T_enhanced = T_base × (0.6 + 0.4 × 7Q_composite)
SYNTHESIS — CAPS
Weak Dimensions Cap the Total
You cannot game this system.
If explanatory depth is missing, the score is capped regardless of evidence strength. If kill conditions are absent, the score is capped regardless of everything else. Three hard caps enforce completeness.
Hard Caps
ED < 0.3 → max(T) = 0.70 (no explanation)
Q7 < 0.3 → max(T) = 0.60 (unfalsifiable)
Q6 < 0.3 → max(T) = 0.70 (no consequences)
Q7 < 0.3 → max(T) = 0.60 (unfalsifiable)
Q6 < 0.3 → max(T) = 0.70 (no consequences)
Selected Node Detail
Click any axiom card for detail.
—
Claim / Function
Select an axiom card above to see its full claim, function within the framework, and worked example.
Formal
—
No trace active. Select a trace example in the right sidebar to highlight how a specific claim flows through all seven questions.