Skip to content

Should "Event" Be a Subtype of "Work"? An Ontological Analysis

Recommendation: Keep Event as a Separate Primitive

Section titled “Recommendation: Keep Event as a Separate Primitive”

Every major formal ontology, knowledge graph, and metadata standard maintains events and works as categorially distinct. The distinction (endurant/perdurant, continuant/occurrent) is considered foundational. No major ontology makes events a subtype of works.

1. Major Ontologies — Unanimous Separation

Section titled “1. Major Ontologies — Unanimous Separation”
OntologyEventsWorks/ObjectsRelationship
BFOOccurrentContinuantDisjoint — separate mereological relations
DOLCEPerdurantEndurantDisjoint categories
SUMOProcessObjectSibling categories under Physical
UFOPerdurant (UFO-B)Endurant (UFO-A)Required UFO-AB to relate them
CIDOC-CRME5 Event (Temporal Entity)E71 Human-Made Thing (Persistent Item)Separate branches of hierarchy
Schema.orgEvent (under Thing)CreativeWork (under Thing)Sibling types
Dublin Core”Non-persistent, time-based occurrence”Persistent resourcesDistinct categories
Wikidatasubclass of “occurrence”subclass of “intellectual work”Separate hierarchies
IFLA-LRMNot modeledWork → Expression → Manifestation → ItemEvents outside FRBR scope

Creation (E65) is an EVENT that PRODUCES a Thing (E71). The creation event and the created thing are distinct:

  • The symphony premiere (event) ≠ the symphony (work)
  • The premiere happened once. The symphony persists.
  • Events have causal coherence. Works have attributive coherence.

ἔργον covers both deed (activity) and product (creation). Aristotle distinguished these senses. This supports having BOTH primitives, not collapsing them.

  • Participants (co-present during unfolding) vs. creators/audiences
  • Non-persistence (happened, now over) vs. persistence
  • Temporal parts (beginning, middle, end that unfold) vs. structural parts that coexist
  • Causal coherence vs. attributive coherence
  • Occurrence semantics (“it happened”) vs. expression semantics (“it was published”)
  • FRBR chain (express_as, adapt, translate, cover) makes no sense for events

5. The Philosophical Dissolution Tradition

Section titled “5. The Philosophical Dissolution Tradition”

Whitehead, Quine, Goodman, Lewis dissolve the distinction — but by making EVERYTHING events/processes, not by making events a subtype of objects. This isn’t useful for a knowledge graph.

Event management uses “production” language, but also recognizes events are ephemeral experiences, not persistent artifacts. The Experience Economy positions events alongside goods and services — distinct categories.

  • Stanford Encyclopedia of Philosophy, “Events” (2025 revision)
  • CIDOC-CRM v7.1.3 (2024)
  • BFO 2.0 / ISO 21838-2
  • DOLCE (Laboratory for Applied Ontology)
  • SUMO (Ontology Portal)
  • UFO (Guizzardi et al.)
  • Schema.org type hierarchy
  • Dublin Core DCMI Type Vocabulary
  • IFLA-LRM (2017)
  • Wikidata Q1656682, Q17537576
  • Stout, “The Category of Occurrent Continuants”
  • Baratella, “Are There Occurrent Continuants? A Reply”
  • Guarino, Baratella & Guizzardi, “Events, their names, and their synchronic structure”