Skip to content

R0021/2026-03-25/Q008 — Assessment

BLUF

The word "set" has 430 definitions in the OED Second Edition (580 senses including phrasal verbs), though "run" has since taken the record with 645 senses. Polysemy is pervasive in natural language — most content words are polysemous, and more frequent words tend to be more polysemous. This stands in stark contrast to formal specification languages where each term has exactly one meaning. Academic research confirms polysemy is "notoriously difficult to treat both theoretically and empirically." For prompt engineering, this means natural language prompts are inherently ambiguous specification instruments.

Probability

Rating: Almost certain (95-99%)

Confidence in assessment: High

Confidence rationale: OED data is verifiable. Polysemy research is peer-reviewed.

Reasoning Chain

  1. "Set" has 430 definitions (OED2) / 580 senses including phrasal verbs [SRC01-E01, Medium reliability, High relevance]
  2. "Run" now holds the record with 645 senses (OED3, 2011) [SRC01-E01]
  3. Polysemy is pervasive: most content words are polysemous; frequency correlates with polysemy [SRC02-E01, High reliability, High relevance]
  4. Formal specification languages assign exactly one meaning per term within scope
  5. JUDGMENT: The ambiguity gap is quantifiable — approximately 430:1 for a common word. This is not a matter of degree but of kind: natural language is fundamentally ambiguous while formal languages are fundamentally precise.

Evidence Base Summary

Source Description Reliability Relevance Key Finding
SRC01 OED "set" data Medium High 430 definitions (OED2); 645 for "run" (OED3)
SRC02 Polysemy research High High Pervasive, frequent words more polysemous, "notoriously difficult"

Collection Synthesis

Dimension Assessment
Evidence quality Medium-High — OED data via secondary source, polysemy research peer-reviewed
Source agreement High — all sources confirm pervasive natural language ambiguity
Source independence Independent — linguistic data and computational linguistics research
Outliers None

Gaps

Missing Evidence Impact on Assessment
Direct comparison with specific formal languages (Z, TLA+) Minor — the 1:many vs 1:1 distinction is well-established
LLM behavior with polysemous prompts Moderate — would quantify impact on prompt engineering

Researcher Bias Check

Declared biases: Researcher argues prompt engineering uses an inherently imprecise tool (natural language). This evidence supports that argument.

Influence assessment: The OED data and polysemy research are independently verifiable. The comparison to formal languages is the researcher's frame.

Cross-References

Entity ID File
Hypotheses H1, H2, H3 hypotheses/
Sources SRC01, SRC02 sources/
ACH Matrix ach-matrix.md
Self-Audit self-audit.md