Sci - Semantic Substrates of Applied Mathematical Notation: A Cross-Disciplinary Survey of Greek Letter Usage, Historical Origin, and Namespace Collision from Classical Analysis to Machine Learning

This paper investigates whether Greek letter choices in mathematics, computer science, machine learning, and field physics follow systematic patterns or are arbitrary conventions in the Saussurean sense. We introduce the concept of the substrate -- the intrinsic behavioral class a symbol encodes at its moment of first significant use -- and demonstrate through a systematic Origin Record spanning sixty distinct symbol uses (Cauchy 1847 to Kingma-Ba 2014) that substrates persist across time, domain changes, and community boundaries at rates significantly exceeding chance. Four empirical findings are derived: (1) Substrate persistence -- substrate-informed assignments produce consistent overloads at a 3:1 ratio; (2) The macro/micro case split -- uppercase encodes aggregate/structural roles, lowercase encodes instance/operational roles; (3) The ML-physics boundary as maximum-density collision zone -- 37% of all breaking pairs involve the post-1986 ML stratigraphy layer in less than 20% of the time window; (4) P-mnemonic origin correlates with adoption success. The paper documents 43 substrate-breaking symbol pairs with severity ratings, 9 substrate-consistent overload families, 82 numbered discoveries, a complete collision registry with primary-source evidence, six navigation strategies ordered by cost, three design rules, and a formal notation declaration for the CyberAxis framework. Published under CC BY-NC 4.0.