# KEHOME/knowledge/theory/NaturalSemanticMetalanguage/happen.html # Oct/29/2003
From my perspective, HAPPEN [1] is the most unusual semantic prime, and the most difficult to map into MKR [2]. HAPPEN deliberately suppresses the agent/actor information which MKR strives to preserve.
To represent NSM HAPPEN, I have chosen to use the MKR command syntax with "do" replaced by "happens".
In the MKR language, entity actions/events are characterized by propositions of the form
at space=s, time=t, view=v { entity do action = event with action characteristics out action products od action direct objects from action initial characteristics to action final characteristics done; };Event is the occurrence/instance of an action. If "= event" is omitted, a unique event name is automatically generated. The NSM concept HAPPEN corresponds to
at space=s, time=t, view=v { entity do action = event od direct object done; };with entity omitted. The new MKR syntax is
at space=s, time=t, view=v { happens action = event od direct object done; };
Although Wierzbicka describes HAPPEN in terms of a single action/event, I think that the concept is intended to apply to interactions. In other words, HAPPEN applies to a sequence of related actions/events. In MKR, interactions are characterized by a "script" of actions/events. The script does not have a single entity as subject.
The only interaction which is explicitly represented in the current version of MKR/MKE is causality.
event:1 causes event:2; or event:2 because event:1;An event may be expressed as an action proposition.
For general interactions, the new syntax is the same as for actions.
happens interaction ... done;