Work

When Does a Field Emerge, and When Does It Matter Causally? A Bidirectional Research Programme Across Physical, Biological, and Language–Signal Systems

Year: 2026
Type: preprint
Abstract: The recurrence of field descriptions across physical, biological, mathematical, and computational sciences creates a methodological problem rather than evidence of a universal ontology. When may a distributed system be said to enter a distinct collective regime? When does an identified field or distributed mechanism participate causally in an observed outcome? And when may a result or technique developed in one domain be transferred to another without transporting more explanatory authority than the evidence permits? This paper proposes a domain-sensitive, bidirectional research method that evaluates these questions separately through regime qualification, causal-participation qualification, and correspondence-and-transfer qualification. It relates established research in non-equilibrium transitions, active matter, bioelectric signaling, neural fields, causal abstraction, transportability, and multiagent systems to candidate Large Language Field (LLF) and Large Signal Field (LSF) architectures. In the forward direction, disciplinary methods constrain computational field hypotheses; in the reverse direction, field-structured representations are assessed for possible incremental gains in scientific evidence traceability, comparison, and qualification. A laboratory extension contract allows independent investigations, including research-book reports, to reuse the foundational method without requiring a separate methodological paper for each study. The proposal specifies observation and intervention requirements, strong information-equivalent baselines, asymmetrical transfer tests, abstention outcomes, versioned evidence, and prospective falsifiers. It reports no new experiment, no discovery of an endogenous computational field, no new causal identification theorem, and no advantage over established methods. Its contribution is a falsifiable, reusable research architecture whose value must itself be established through independent, domain-appropriate assessment.
Language: English
Open Access status: green
OpenAlex ID: W7225871896
Cites:
Cited by:
Related to: 10