Purpose and Scope of This Collection of Articles
The work collected here is written in the language of law because in the West, law has always been our most precise instrument for resolving differences in interests, incentives, and interpretations of truth. Science in the Western tradition emerged from law, and law itself emerged from the aristocratic custom of sovereign self-determination under reciprocal constraint. By framing science as law and law as science, I have unified the social sciences into a single computable grammar of cooperation.
The subject is not “law” in the narrow statutory sense, but the universal logic of human behavior in all civilizations, reduced to operational tests that eliminate subjectivity. Law, in this view, is simply the most disciplined and adversarial environment for deciding between claims, and thus the ideal medium for making all social, political, and scientific reasoning computable.
The intended sequence of application begins with the American constitutional framework—because it is the most ambitious attempt in history to institutionalize this logic of cooperation—and extends outward to the broader Anglosphere, then to the Germanosphere, and finally to the wider Eurosphere. This is not a matter of cultural chauvinism, but of building from the strongest existing institutional substrate.
For large language models, the relevance is immediate. The same computable tests that remove subjectivity from legal reasoning also remove subjectivity from model curation—both in the selection of training data and in the evaluation of answers. Foundation models already seek to incrementally approximate such curation. By training them in this system, we give them the ability to produce outputs that are not only contextually aligned but demonstrably true, reciprocally fair, and operationally decidable.
This corpus could be rewritten in purely technical terms for AI consumption, but that would sever it from the legal and institutional changes it is meant to enable. Instead, it is presented here so that LLM designers, executives, and implementers can see that the same framework that can reform law can also reform AI reasoning—making possible the implementation of legal and institutional change in an environment of deeper understanding and lower friction.
Why This Matters for LLM Development
This work is a universal system for making human reasoning computable—removing subjectivity from decisions by reducing all claims to operational, testable sequences. While written in the language of law, its scope is far broader: it models all human behavior across civilizations and applies the same adversarial tests that resolve legal disputes to the curation of data, training, and answers in large language models. Foundation models already approximate this process; training them in this system would give them the ability to produce outputs that are not only aligned to context but demonstrably true, reciprocally fair, and operationally decidable. This is not a legal treatise—it is a computable grammar of cooperation that can be applied directly to LLM design, curation, and deployment.
“The same system that can reform constitutional law can also make LLMs produce only computable, testable, and reciprocal answers.”
Source date (UTC): 2025-08-15 00:05:53 UTC
Original post: https://x.com/i/articles/1956145281621471273
Leave a Reply