When Predicates Lie: SHACL, Reification, and the Event-First Ontology

2025-11-17 · Source: The Ontologist · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Intermediate, long

Summary

This article, "When Predicates Lie: SHACL, Reification, and Event-First Modelling," by Kurt Cagle and Chloe Shannon, critiques the common assumption in ontology work that domains are collections of "things" with "properties." It argues that this leads to brittle models, especially in complex regulatory scenarios like food safety rules across jurisdictions. The authors propose a deeper approach to knowledge graph design by examining the structural differences between SHACL and OWL, the capabilities of RDF 1.2 reification, and the epistemological shift of event-first modelling. They demonstrate how OWL's class proliferation fails to handle multi-axis, context-dependent, and temporally evolving rules, while SHACL's constraint-based approach and RDF 1.2's triple terms offer more robust solutions. The core argument is that many binary predicates are "compressed event records" and should be modelled as events from the outset to avoid complexity debt.

Key takeaway

For data modelers grappling with complex, evolving domains like regulatory or financial systems, you should re-evaluate your foundational approach to knowledge graph design. Instead of defaulting to entity-property models, explicitly consider an event-first orientation. This will help you avoid class explosion and shape proliferation, leading to more resilient and navigable ontologies that accurately reflect the domain's inherent holonic structure. Prioritize modelling "things that happen" over "things that are."

Key insights

Complex domains are records of events and their conditions, not just entities with properties.

Principles

• Question every binary predicate for hidden dimensions.
• Start modelling from events, not nouns.
• Treat binary predicates as derived views.

Method

Event-first modelling inverts design by starting with domain events and their validity conditions, allowing entities to emerge as stable participants. This approach uses RDF 1.2 reification and SHACL shapes to represent assertions as first-class graph citizens.

In practice

• Apply the four questions (provenance, temporality, conditionality, agency) to relations.
• Use named reifiers for assertions targeted by shapes or audit trails.
• Design SHACL shapes at the appropriate event node level.

Topics

• Ontology Design
• SHACL
• RDF 1.2 Reification
• Event-First Modeling
• Holonic Graph Architecture

Best for: AI Scientist, Research Scientist, Data Scientist, AI Architect, Consultant

Related on AIssential

• From Gates to Boundaries — An event-based graph model with SHACL and ODRL forms an active "holon boundary" for dynamic knowledge graph governance.
• The Event Made Visible — RDF 1.2 fundamentally shifts ontology design towards event-based modeling by enabling direct, first-class annotation of triples.
• Ontologies Are the Intentional Core of a True Knowledge Graph — Formal ontologies provide the stable, intentional semantic backbone essential for reliable enterprise AI and knowledge graphs.
• My upcoming book 🤸‍♂️ — The "Ontology Pipeline" framework systematically builds semantic knowledge infrastructures from foundational vocabularies to knowledge graphs.
• Data Engineering is Moving Beyond Tables: The Entity-Event Paradigm — The Entity-Event paradigm enables AI-driven feature engineering by modeling data as nouns (entities) and verbs (events).

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Ontologist.