If languages are ontologies — structured potentials for construing experience — then typology is not a catalogue of structural curiosities. It is the comparative study of how different systems cut experience into events, participants, and relations.
This post begins the deeper dive by examining the most fundamental of all ontological distinctions in the ideational metafunction:
What counts as an event?What counts as a participant?And how different languages draw that boundary.
These distinctions are usually treated as grammatical preferences, or “encoding choices”, or “argument structure patterns”. In a relational ontology, we treat them instead as affordances for construing phenomena, where:
-
the system (language) = a theory of possible construals;
-
the instance (text) = an event of selecting a particular way of cutting experience;
-
meaning = reality in the relational sense — not a mirror of an unconstrued world, but the very act of bringing a phenomenon into structured presence.
1. The Event–Participant Boundary: Languages Disagree About What Happens
In English (and many Indo-European languages), the event boundary is drawn around verbs. Verbs are where change, doing, being, and happening live. Participants (nouns) are relatively inert, and their involvement in the event is mediated by the verb.
This yields a construal where:
-
events are located in verbal clauses,
-
participants “fill roles” inside the event,
-
and relationality is syntactically centralised in the verb.
But that is only one way to cut experience.
Languages with Participant-centric event construals
Many languages — from Salish to certain Austronesian systems — distribute eventhood across the clause in ways that make participants more eventful and events less verb-centric.
In some languages:
-
nouns are morphologically marked for control, affectedness, or dynamism,
-
verb-like properties (aspect, valence, animacy-relations) appear on nominal elements,
-
the line between “actor” and “event” is shaded rather than sharp.
This does not mean such languages “lack verbs” (a tired myth), but that their system encodes event relations differently — as distributed properties rather than centralised verbal nuclei.
2. Ontological Consequence 1: Different Languages Offer Different Theories of Causation
When the event boundary shifts, so does the construal of causation.
Verb-centric languages (English, French, Hindi, etc.)
Causation is usually a structured relation between:
-
a causer
-
a caused event
-
sometimes an affected participant
This makes causation appear:
-
hierarchical,
-
agentive,
-
often unidirectional.
Participant-centric or distributed-event languages
Causation can be construed as:
-
emergent from participant configurations,
-
distributed rather than initiated,
-
sometimes reciprocal or non-linear.
A speaker of such a language does not “think differently because of grammar” — that would slip back into determinism — but their system affords different event-cuts, different metaphenomenal categories, and different stories of what it means for something to cause something else.
3. Ontological Consequence 2: Agency Is Not Universal — It Is a Construal Option
English habitually assigns agency to animate participants, often even when no agent exists:
-
The wind blew the door open.
-
My computer ate the file.
Other languages avoid this pattern entirely, construing such phenomena as:
-
event-internal changes,
-
state shifts without external instigators,
-
or configurations in which no “agent” is needed.
This does not reflect a difference in metaphysics. It reflects a difference in construal affordances:
-
some systems normalise agentive stories,
-
others normalise non-agentive ontologies.
Agency becomes a linguistic choice, not a universal category of experience.
4. Ontological Consequence 3: Some Languages Don’t Separate Events from States
In English we sharply distinguish:
-
events (run, explode, melt)
-
states (know, belong, love)
But other languages cut the continuum differently:
-
Some treat states as extended events with internal temporality.
-
Some fuse possession, location, and state as a single relational domain.
-
Some encode state-change with the same morphology as agentive action.
This matters because the state/event division is often taken as an ontological universal. It is not. It is a systemic construal option encoded by some languages more rigidly than others.
5. So What Is the Big Claim?
The relational-ontological claim is simpler, cleaner, and far more powerful:
Different languages instantiate different theories of how experience can be cut into events, participants, and relations.These cuts are not windows onto an unconstrued world.They are how phenomena become phenomena at all.
And because construal = meaning = reality (in first-order terms), typology becomes:
-
the study of contrasting ontologies,
-
not as metaphysical commitments,
-
but as semiotic architectures of the possible.
No comments:
Post a Comment