Computational Detection of Constitutional Drift: Network Analysis and Semantic Measurement of Argentine Supreme Court Jurisprudence (1922–2025)

This study introduces a computational framework for quantifying constitutional degradation through judicial citation network analysis and semantic text mining. Traditional legal scholarship relies on qualitative interpretation, which limits systematic cross-jurisdictional comparison and predictive modeling of constitutional evolution. 

This paper addresses that gap by developing three open-source algorithms applied to 72 Argentine Supreme Court decisions (1989-2025). 

Three algorithms were developed: JurisRank (a PageRank adaptation measuring doctrinal fitness through citation networks), RootFinder (Ancestral Borrowing Analysis Network for genealogical concept tracing), and Legal-Memespace (principal component analysis for multidimensional doctrine mapping).

Cases were selected through stratified sampling covering emergency decrees, economic regulations, and control domains, with inter-coder reliability of κ=0.83 (95% CI: 0.76–0.89). Formalist constitutional interpretations declined in network fitness from 0.89 (1922 baseline) to 0.03 (2025), a reduction of 97%, while emergency doctrines rose from 0.11 to 0.94 (Kendall’s τ = −0.89, p<0.001). K-fold cross-validation (k=5) yields mean accuracy of 73.2%. In the 2024–2025 period, institutional resistance to executive decrees correlated 0.91 with fiscal impact across 28 analyzed measures, suggesting selective constitutional enforcement. Monte Carlo simulations (n=1,000) assign 82% probability to formalist doctrine reaching functional extinction (fitness <0.05) by 2030.

The framework enables automated constitutional monitoring, cross-jurisdictional comparison, and early crisis detection.

All algorithms are released as open-source software (github.com/adrianlerer/peralta-metamorphosis) for independent validation and collaborative extension, demonstrating that evolutionary legal theory can transition from speculative framework to empirical research program.