Can explicit linguistic measures add something to AI-based developmental interpretation?
I reanalysed the El-Sayed–Rogers debate using our Transcendental Pluralism assessment prototype: a parser, explicit rules for grouping meaningful units, and abstraction heuristics.
Integral Life’s Debate Analyzer places El-Sayed usually at 3.0, Rogers between 2.5 and 3.0, and both reaching 3.5. In its cross-framework terminology, 3.0 corresponds to STAGES Expert—reasoning within a governing framework—and 3.5 to Achiever, coordinating multiple variables toward outcomes.
Our assessment found recurring 3.3 Systematic structures in both speakers: organized groups of meaning coordinated within a larger whole. El-Sayed also produced one passage scored at 3.4 Metasystematic, organizing relationships between systems.
Under our proposed correspondences, these indicate later structural complexity than Expert–Achiever. The numbers belong to different scales: 3.3 here is not directly comparable to 3.0 there. “Developmental altitude” means an approximate level of complexity; correspondence between stage descriptions is not an established statistical correlation.
We assess the structure expressed in utterances, rather than infer someone’s overall maturity from their political positions. Highest observed complexity also differs from typical performance.
This remains exploratory: a human scorer checked the accuracy of our program that currently is limited to the 3.4 Stage - in itself a far to early end point to make it generally applicable.
But the question is: how can approaches like this shape future political decisions - simply analyzing complexity won't be enough, analyzing political positions and grading their stage, likely neither - what would be an integrated approach from your perspective?