Skip to content
Needs Verification

Degeneration Index & Immunization Tracking

Definition & Conceptual Goal

The Degeneration Index & Immunization Tracking metric suite measures the Lakatosian degeneration of a scientific research program across historical theory versions and detects ad-hoc immunizing stratagems (Balzer et al., 1987, pp. 333–338; Stegmüller, 1976, p. 184; Schurz, 2024, pp. 51, 59, 61, 265).

It diagnoses two interconnected layers of non-progressive theory evolution:

  1. Micro-Level (Node Immunization): Detects when a new auxiliary hypothesis or core expansion (\(K_{i+1}\)) absorbs an empirical anomaly (\(a \in I\)) but generates zero independent excess empirical content (\(\operatorname{Cn}(K_{i+1}) \setminus \{a\} \subseteq \operatorname{Cn}(K_i)\)).
  2. Macro-Level (Programme Degeneration Index \(D\)): Quantifies the ratio of ad-hoc resolved anomalies (Typ-b failures) to the sum of genuine empirical successes and honest direct empirical refutations (Typ-a failures).

Theoretical Grounding & Model Formulation

In Schurz's reconstruction of Lakatos's methodology of scientific research programmes: For each historical version \(V_j\) of a theory:

  • \(S (V_j)\) (Genuine Successes): Phenomena successfully predicted or explained without ad-hoc adjustments.
  • \(F_{\text{Typ a}} (V_j)\) (Direct Failures): Well-established empirical phenomena that directly contradict \(V_j\) logically or probabilistically.
  • \(F_{\text{Typ b}} (V_j)\) (Immunized / Ad-Hoc Failures): Phenomena that contradicted a prior version \(V_{j-1}\), but were absorbed in \(V_j\) via auxiliary hypotheses lacking independent empirical confirmation.

An anomaly resolution \(E_{i+1} = \langle K_{i+1}, I_{i+1} \rangle\) is an ad-hoc immunization if: $$ \Delta_{\text{anom}} > 0 \quad \text{and} \quad \Delta_{\text{content}} = |\operatorname{Cn} (K_{i+1}) \setminus \operatorname{Cn} (K_i)| = 0 $$

Mathematical Specification & Graph Formulation

Node-Level Immunization Index (\(II\))

For an auxiliary modification introducing new laws resolving anomalies:

\[II (K_i, K_{i+1}) = \begin{cases} 1.0 & \text{if } \Delta_{\text{anom}} > 0 \land \Delta_{\text{content}} = 0 \\ \frac{\Delta_{\text{anom}}}{\Delta_{\text{anom}} + \Delta_{\text{content}}} & \text{otherwise} \end{cases}\]

Ratio of Progressive to Immunizing Nodes (\(RPI\))

Across the entire evolution sequence \(\mathcal{E}\):

\[RPI (\mathcal{E}) = \frac{|V_{\text{progressive}}|}{|V_{\text{progressive}}| + |V_{\text{immunizing}}|}\]

Schurz's Degeneration Index (\(D\))

For historical theory version \(V_j\):

\[D (V_j) = \frac{|F_{\text{Typ b}} (V_j)|}{|S (V_j)| + |F_{\text{Typ a}} (V_j)|}\]

Programme Trajectory Trend (\(\Delta D\))

Over a sequence of versions \(V_1, V_2, \dots, V_m\):

\[\Delta D = \frac{D (V_m) - D (V_1)}{m - 1}\]
  • \(\Delta D > 0\): Degenerating Research Programme (ad-hoc immunizations accumulate faster than real progress).
  • \(\Delta D \le 0\): Progressive Research Programme (genuine successes outpace anomalies).

Measurement & Graph Implementation

  1. Detect Ad-Hoc Nodes:
  2. Trace incoming :RESOLVES_ANOMALY edges connected to new auxiliary hypothesis nodes.
  3. Query for outgoing :PREDICTS_NOVEL edges. Auxiliary nodes with incoming anomaly resolutions but zero novel predictions are labeled :AdHocImmunization (Typ-b failure).
  4. Compute Version Degeneration (\(D\)): 
    MATCH (v:TheoryVersion {id: $version_id})
OPTIONAL MATCH (v)-[:HAS_SUCCESS]->(s:Success)
OPTIONAL MATCH (v)-[:HAS_FAILURE_A]->(fa:FailureTypA)
OPTIONAL MATCH (v)-[:HAS_FAILURE_B]->(fb:FailureTypB)
WITH count(DISTINCT s) AS s_count, count(DISTINCT fa) AS fa_count, count(DISTINCT fb) AS fb_count
RETURN fb_count * 1.0 / (s_count + fa_count + 1e-6) AS degeneration_index

Diagnostic & Metascientific Value

Measurement Indicator Programme Dynamic Metascientific Interpretation
\(\Delta D \le 0, D < 0.3\) Progressive Programme Healthy Lakatosian growth; true empirical successes dominate.
\(\Delta D \approx 0, D \sim 0.5\) Stagnant Programme Defensive maintenance; new predictions roughly balance anomaly patches.
\(\Delta D > 0, D > 1.0\) Degenerating Programme Defensive crisis; majority of modifications are ad-hoc immunizations.
\(II = 1.0\) (Node Level) Pure Immunizing Patch Auxiliary node shields theory from falsification without advancing science.

Grounding References

  • [Schurz, 2024] Schurz, G. (2024). Philosophy of Science: A Unified Approach. Routledge, pp. 51, 59, 61, 265.
  • [Balzer et al., 1987] Balzer, W., Moulines, C. U., & Sneed, J. D. (1987). An Architectonic for Science. Reidel Publishing, pp. 333–338.
  • [Stegmüller, 1976] Stegmüller, W. (1976). The Structure and Dynamics of Theories. Springer-Verlag, p. 184.