Aziel's Razor and Occam's Razor: A Variable-Completeness Framework for Model Construction Before Parsimony

Abstract

Aziel's Razor and Occam's Razor A variable-completeness framework for model construction before parsimony Occam’s Razor is useful as a consolidation tool. It is frequently misused as a discovery filter. In practice, investigators often simplify before they have mapped the relevant variables. The resulting models look clean only because material factors were dropped early. This paper names that sequencing error and proposes a corrective rule. Aziel’s Razor states: no materially plausible, evidence-linked variable should be discarded merely because it complicates the explanation. Investigators should first enumerate candidate variables, preserve their provenance, test dependencies, and only then consolidate to the smallest model that retains observed structure and declared uncertainty. The framework does not reject parsimony. It reorders it. Parsimony becomes the final compression step rather than the initial filter. Contents:- Formal definition and operating principles- Direct comparison with common formulations of Occam’s Razor- Eight-step operational protocol- Material plausibility criteria to prevent unlimited expansion- Failure modes and safeguards- Five testable propositions- Proposed comparative evaluation design- Explicit limitations and status note Status: Conceptual framework. Not yet independently validated. Prepared for scholarly critique. Core instruction: do not simplify before understanding what is being simplified.