Multiword predicates
are predicates that consist of multiple words. They include light verb
constructions, which we annotate with the META frame. Other
dependents that do not fill a role of the head predicate but rather are part of
the predicate – as in verbal idioms, weather verbs, existential there, names,
other fixed expressions, or specialized sublanguages for such things as dates
and times – are marked with NONCOMP:
Dependency graph for sentence: Kim kicked the bucket. Token kicked is labeled EXISTENCE-DEINIT and has an edge labeled existent to token Kim and an edge labeled NONCOMP to token bucket.Dependency graph for sentence: It is raining. Token raining is labeled ENTITY and has an edge labeled NONCOMP to token It.Dependency graph for sentence: There was a famine. Token was is labeled META and has an edge labeled NONCOMP to token There and an edge labeled scene to token famine.Dependency graph for sentence: the asteroid known as B 612. Token 612 is labeled ENTITY and has an edge labeled NONCOMP to token B.Dependency graph for sentence: One drawing goes along all right. Token goes is labeled SITUATION-DYN and has an edge labeled situee to token drawing and an edge labeled NONCOMP to token along and an edge labeled SCENE to token right.Dependency graph for sentence: “ It is a question of discipline , ” the little prince said to me later on. Token on is labeled TIME and has an edge labeled has-time to token said and an edge labeled NONCOMP to token later.
In some cases, the arguments of a multiword predicate depend on different
components. If the component heads a core argument of the multiword predicate,
give it the same frame as the head component. Otherwise, use ENTITY.
Dependency graph for sentence: I had the sheep to thank for it. Token had is labeled CAUSATION and has an edge labeled EXPERIENCER to token I and an edge labeled causer to token sheep and an edge labeled NONCOMP to token thank. Token thank is labeled CAUSATION and has an edge labeled effect to token it.Dependency graph for sentence: I jumped to my feet. Token jumped is labeled POSE-INIT and has an edge labeled has-pose to token I and an edge labeled has-pose to token my and an edge labeled NONCOMP to token feet. Token feet is labeled POSE-INIT.Dependency graph for sentence: Are you pulling my leg. Token pulling is labeled SENDING-INIT and has an edge labeled sender to token you and an edge labeled NONCOMP to token leg. Token leg is labeled ENTITY and has an edge labeled EXPERIENCER to token my.
Bound pronouns that are part of multiword predicates (e.g., inherently
reflexive verbs) are annotated with the same role as their antecedent:
Dependency graph for sentence: Kim erinnert sich nicht daran. Token erinnert is labeled MESSAGE and has an edge labeled EXPERIENCER to token Kim and an edge labeled EXPERIENCER to token sich and an edge labeled SCENE to token nicht and an edge labeled topic to token daran. Token nicht is labeled NEG and has an edge labeled negated to token erinnert.Dependency graph for sentence: I address myself to you. Token address is labeled SENDING-INIT and has an edge labeled sender to token I and an edge labeled sender to token myself and an edge labeled EXPERIENCER to token you.Dependency graph for sentence: I had with me neither a mechanic nor any passenger. Token had is labeled ACCOMPANIMENT and has an edge labeled accompanied to token I and an edge labeled accompanied to token me and an edge labeled accompanier to token mechanic and an edge labeled accompanier to token passenger.
Occasionally, there is no (direct) syntactic connection between the components
of a multiword predicate. For example as … as … could be analyzed as a
predicate with two components. We do not currently have a way to mark this, so
we just annotate the parts as if each part were a full predicate.
Dependency graph for sentence: as soon as possible. Token as is labeled COMPARISON and has an edge labeled compared to token soon. Token as is labeled COMPARISON and has an edge labeled reference to token possible.
Dependency graph for sentence: Kim kicked the bucket. Token kicked is labeled EXISTENCE-DEINIT and has an edge labeled existent to token Kim and an edge labeled NONCOMP to token bucket.
Dependency graph for sentence: It is raining. Token raining is labeled ENTITY and has an edge labeled NONCOMP to token It.
Dependency graph for sentence: There was a famine. Token was is labeled META and has an edge labeled NONCOMP to token There and an edge labeled scene to token famine.
Dependency graph for sentence: the asteroid known as B 612. Token 612 is labeled ENTITY and has an edge labeled NONCOMP to token B.
Dependency graph for sentence: One drawing goes along all right. Token goes is labeled SITUATION-DYN and has an edge labeled situee to token drawing and an edge labeled NONCOMP to token along and an edge labeled SCENE to token right.
Dependency graph for sentence: “ It is a question of discipline , ” the little prince said to me later on. Token on is labeled TIME and has an edge labeled has-time to token said and an edge labeled NONCOMP to token later.
Dependency graph for sentence: I had the sheep to thank for it. Token had is labeled CAUSATION and has an edge labeled EXPERIENCER to token I and an edge labeled causer to token sheep and an edge labeled NONCOMP to token thank. Token thank is labeled CAUSATION and has an edge labeled effect to token it.
Dependency graph for sentence: I jumped to my feet. Token jumped is labeled POSE-INIT and has an edge labeled has-pose to token I and an edge labeled has-pose to token my and an edge labeled NONCOMP to token feet. Token feet is labeled POSE-INIT.
Dependency graph for sentence: Are you pulling my leg. Token pulling is labeled SENDING-INIT and has an edge labeled sender to token you and an edge labeled NONCOMP to token leg. Token leg is labeled ENTITY and has an edge labeled EXPERIENCER to token my.
Dependency graph for sentence: Kim erinnert sich nicht daran. Token erinnert is labeled MESSAGE and has an edge labeled EXPERIENCER to token Kim and an edge labeled EXPERIENCER to token sich and an edge labeled SCENE to token nicht and an edge labeled topic to token daran. Token nicht is labeled NEG and has an edge labeled negated to token erinnert.
Dependency graph for sentence: I address myself to you. Token address is labeled SENDING-INIT and has an edge labeled sender to token I and an edge labeled sender to token myself and an edge labeled EXPERIENCER to token you.
Dependency graph for sentence: I had with me neither a mechanic nor any passenger. Token had is labeled ACCOMPANIMENT and has an edge labeled accompanied to token I and an edge labeled accompanied to token me and an edge labeled accompanier to token mechanic and an edge labeled accompanier to token passenger.
Dependency graph for sentence: as soon as possible. Token as is labeled COMPARISON and has an edge labeled compared to token soon. Token as is labeled COMPARISON and has an edge labeled reference to token possible.