Universal Relations

Universal Relations, formerly known as "links", are labelled arcs connecting a node to another node in a UNL graph. They correspond to two-place semantic predicates holding between two Universal Words. In UNL, universal relations have been normally used to represent semantic cases or thematic roles (such as agent, object, instrument, etc.) between UWs. The repertoire of universal relations is defined in the UNL Specs and it is not open to frequent additions.

Definition

In the UNL framework, universal relations describe semantic functions between two UWs. These functions are binary and directed (from a source to a target) and are claimed to be universal. Because of their similarity in name and function to syntactic relations, it may seem that the labels used for relations are different names for special grammatical functions. This is emphatically not the case. The intention is that the labels used denote specific ideas rather than grammatical structures: the idea of “something that initiates an event,” or “agent” for example, is quite different from “grammatical subject of a sentence”, even though many times the subject of a sentence will indicate the agent of the event. The agent of an event may also appear as an adjective or noun modifier, with the preposition “by” or embedded in nouns with “er” suffixes. The whole point of the conceptual relations is to have a name for these very different grammatical structures which are conceptually quite the same. Thus, the conceptual relations used in UNL are much more abstract than the grammatical relations found in sentences.

Syntax

Universal relations are represented as follows:

<rel>:<scope>(<source>;<target>)

where:

• <rel> is the name of the relation (two-character or three-character lower-case strings) (see the complete list of relations below)
• <scope> is the scope of the relation (two-character unique identifier for the scope). The scope may be omitted if the main scope, i.e., :00. See scope.
• <source> is the UW that assigns the relation <rel>
• <target> is the UW that receives the relation <rel>.

Hierarchy of relations

Universal Relations are organized in a hierarchy where lower nodes subsume upper nodes. The topmost level is the relation "rel", which simply indicates that there is a semantic relation between two elements.

Observations

1. Arguments of relations are not commutative:

   cnt(evidence;absence),i.e., evidence of absence, is different from cnt(absence;evidence), i.e., absence of evidence

2. The target always defines the relation:

   <relation>(<source>;<target>)

   • agt(kill;Peter) = Peter is the agent of kill
   • obj(kill;Peter) = Peter is the patient of kill
   • tim(kill;yesterday) = yesterday is the time of kill
   • plc(kill;kicthen) = kitchen is the place of kill
   • mod(book;beautiful) = beautiful is a modifier of book
   • icl(document;book) = book is a type of document
   • iof(city;Paris) = Paris is an instance of city
   • agt:01(kill;Peter) = Peter is the agent of kill in the scope :01 (see scope).

3. Relations describe semantic dependencies rather than syntactic roles.

   The same relation may play different syntactic roles. Consider, for instance, the case of the relation 'gol' (goal):

   • Specifier: Peter received the book = gol(received;Peter)
   • Complement: Mary gave the book to Peter = gol(gave;Peter)
   • Adjunct: Mary bought a book to Peter = gol(bought;Peter)

4. Lexical and syntactic ambiguities are solved through relations.

   Consider, for instance, the case of the English preposition "in", as in Peter works in X.

   • plc(work;X) = X is the physical place where Peter works (the same as Peter works in Geneva)
   • lpl(work;X) = X is the logical place where Peter works (the same as Peter works in politics)
   • cnt(work;X) = X is the content of the work of Peter (the same as Peter works in improving a given technology)
   • tim(work;X) = X is the time when Peter works (the same as Peter works in the summer)
   • dur(work;X) = X is the duration of the work of Peter (the same as Peter works in ten hours)
   • man(work;X) = X is the manner Peter works (the same as Peter works in intervals)

5. Relations are not necessarily bound to a given lexical category.

   The same relation may be used to describe nominal and verbal structures:

   • agt: John arrived = agt(arrived;John), arrival of John = agt(arrival;John)
   • gol: go to NY = gol(go;NY), train to NY = gol(train;NY)
   • cnt: talk about John = cnt(talk;John), book about John = cnt(book;John)

6. In most cases, lower relations may be completely replaced by the corresponding upper levels with the help of attributes:
   • come from NY = src(come;NY) = plc(come;NY.@from)
   • go to NY = gol(go;NY) = plc(go;NY.@to)
   • go through Geneva = via(go;Geneva) = plc(go;Geneva.@through)
   • work since early = tmf(work;early) = tim(work;early.@since)
   • work until late = tmt(work;late) = tim(work;late.@until)
   • work during the summer = dur(work;summer) = tim(work;summer.@during)
   • live and die = seq(live;die) = tim(live;die.@after)
   • kill with a knife = ins(kill;knife) = man(kill;knife.@with)
   • etc.
7. In several cases, however, relations are not completely interchangeable, and replacement implies a significant semantic loss. In these cases, upper levels must be used carefully, and only when there is no other alternative:
   • Peter knows = exp(know;Peter) ≈ agt(know;Peter)
   • Mary is loved = cnt(love;Mary) ≈ obj(love;Mary)
   • A building is created = res(create;building) ≈ obj(create;building)
   • John is Peter = equ(Peter;John) ≈ aoj(Peter;John)
   • Dogs are mammals = icl(mammals;dogs) ≈ aoj(mammals;dogs)
   • John is a human being = iof(human being;John) ≈ aoj(human being;John)
   • table of wood = mat(table;wood) ≈ mod(table;wood)
   • city of New York = nam(city;New York) ≈ mod(city;New York)
   • Peter's book = pos(book;Peter) ≈ mod(book;Peter)

List of relations in alphabetical order

Tag	Relation	Definition	Example
agt	agent	A participant in an action or process that provokes a change of state or location.	John killed Mary = agt(killed;John) Mary was killed by John = agt(killed;John) arrival of John = agt(arrival;John)
and	conjunction	Used to state a conjunction between two entities.	John and Mary = and(John;Mary) both John and Mary = and(John;Mary) neither John nor Mary = and(John;Mary) John as well as Mary = and(John;Mary)
ant	opposition or concession	Used to indicate that two entities do not share the same meaning or reference. Also used to indicate concession.	John is not Peter = ant(Peter;John) 3 + 2 != 6 = ant(6;3+2) Although he's quiet, he's not shy = ant(he's not shy;he's quiet)
aoj	stative object	The subject of an stative verb. Also used to express the predicative relation between the predicate and the subject.	John has two daughters = aoj(have;John) the book belongs to Mary = aoj(belong;book) the book contains many pictures = aoj(contain;book) John is sad = aoj(sad;John) John looks sad = aoj(sad;John);
bas	basis	The basis of a comparison.	John is more beautiful than Peter = bas(beautiful;Peter) John is as intelligent as Mary = bas(intelligent;Mary) John is the most intelligent of us = bas(intelligent;we)
ben	beneficiary	A participant who is advantaged or disadvantaged by an event.	John works for Peter = ben(works;Peter) John gave the book to Mary for Peter = ben(gave;Peter)
cnt	content or theme	The object of an stative or experiental verb, or the theme of an entity.	John has two daughters = cnt(have;two daughters) the book belongs to Mary = cnt(belong;Mary) the book contains many pictures = cnt(contain;many pictures) John believes in Mary = cnt(believe;Mary) John saw Mary = cnt(saw;Mary) John loves Mary = cnt(love;Mary) The explosion was heard by everyone = cnt(hear;explosion) a book about Peter = cnt(book;Peter)
con	condition	A condition of an event.	If I see him, I will tell him = con(I will tell him;I see him) I will tell him if I see him = con(I will tell him;I see him);
dur	duration or co-occurrence	The duration of an entity or event.	John worked for five hours = dur(worked;five hours) John worked hard the whole summer = dur(worked;the whole summer) John completed the task in ten minutes = dur(completed;ten minutes) John was reading while Peter was cooking = dur(John was reading;Peter was cooking)
equ	synonym or paraphrasis	Used to indicate that two entities share the same meaning or reference. Also used to indicate semantic apposition.	The morning star is the evening star = equ(evening star;morning star) 3 + 2 = 5 = equ(5;3+2) UN (United Nations) = equ(UN;United Nations) John, the brother of Mary = equ(John;the brother of Mary)
exp	experiencer	A participant in an action or process who receives a sensory impression or is the locus of an experiential event.	John believes in Mary = exp(believe;John) John saw Mary = exp(saw;John) John loves Mary = exp(love;John) The explosion was heard by everyone = exp(hear;everyone)
fld	field	Used to indicate the semantic domain of an entity.	sentence (linguistics) = fld(sentence;linguistics)
gol	final state, place, destination or recipient	The final state, place, destination or recipient of an entity or event.	John received the book = gol(received;John) John won the prize = gol(won;John) John changed from poor to rich = gol(changed;rich) John gave the book to Mary = gol(gave;Mary) He threw the book at me = gol(threw;me) John goes to NY = plt(go;NY) train to NY = plt(train;NY)
icl	is a kind of	Used to refer to a subclass of a class.	Dogs are mammals = icl(mammal;dogs)
ins	instrument or method	An inanimate entity or method that an agent uses to implement an event. It is the stimulus or immediate physical cause of an event.	The cook cut the cake with a knife = ins(cut;knife) She used a crayon to scribble a note = ins(used;crayon) That window was broken by a hammer = ins(broken;hammer) He solved the problem with a new algorithm = ins(solved;a new algorithm) He solved the problem using an algorithm = ins(solved;using an algorithm) He used Mathematics to solve the problem = ins(used;Mathematics)
iof	is an instance of	Used to refer to an instance or individual element of a class.	John is a human being = iof(human being;John)
lpl	logical place	A non-physical place where an entity or event occurs or a state exists.	John works in politics = lpl(works;politics) John is in love = lpl(John;love) officer in command = lpl(officer;command)
man	manner	Used to indicate how the action, experience or process of an event is carried out.	John bought the car quickly = man(bought;quickly) John bought the car in equal payments = man(bought;in equal payments) John paid in cash = man(paid;in cash) John wrote the letter in German = man(wrote;in German) John wrote the letter in a bad manner = man(wrote;in a bad manner)
mat	material	Used to indicate the material of which an entity is made.	A statue in bronze = mat(statue;bronze) a wood box = mat(box;wood) a glass mug = mat(mug;glass)
mod	modifier	A general modification of an entity.	a beautiful book = mod(book;beautiful) an old book = mod(book;old) a book with 10 pages = mod(book;with 10 pages) a book in hard cover = mod(book;in hard cover) a poem in iambic pentameter = mod(poem;in iambic pentamenter) a man in an overcoat = mod(man;in an overcoat)
nam	name	The name of an entity.	The city of New York = nam(city;New York) my friend Willy = nam(friend;Willy)
obj	patient	A participant in an action or process undergoing a change of state or location.	John killed Mary = obj(killed;Mary) Mary died = obj(died;Mary) The snow melts = obj(melts;snow)
opl	objective place	A place affected by an action or process.	John was hit in the face = opl(hit;face) John fell in the water = opl(fell;water)
or	disjunction	Used to indicate a disjunction between two entities.	John or Mary = or(John;Mary) either John or Mary = or(John;Mary)
per	proportion, rate, distribution or measure	Used to indicate a measure or quantification of an event.	The course was split in two parts = per(split;in two parts) twice a week = per(twice;week) The new coat costs $70 = per(cost;$70)
plc	place	The location or spatial orientation of an entity or event.	John works here = plc(work;here) John works in NY = plc(work;NY) John works in the office = plc(work;office) John is in the office = plc(John;office) a night in Paris = plc(night;Paris)
pof	is part of	Used to refer to a part of a whole.	John is part of the family = pof(family;John)
pos	possessor	The possessor of a thing.	the book of John = pos(book;John) John's book = pos(book;John) his book = pos(book;he)
ptn	partner	A secondary (non-focused) participant in an event.	John fights with Peter = ptn(fight;Peter) John wrote the letter with Peter = ptn(wrote;Peter) John lives with Peter = ptn(live;Peter)
pur	purpose	The purpose of an entity or event.	John left early in order to arrive early = pur(John left early;arrive early) You should come to see us = pur(you should come;see us) book for children = pur(book;children)
qua	quantity	Used to express the quantity of an entity.	two books = qua(book;2) a group of students = qua(students;group)
res	result or factitive	A referent that results from an entity or event.	The cook bake a cake = res(bake;cake) They built a very nice building = res(built;a very nice building)
rsn	reason	The reason of an entity or event.	John left because it was late = rsn(John left;it was late) John killed Mary because of John = rsn(killed;John)
seq	sequence or consequence	Used to express a sequence of events.	John cleaned the room before Peter left = seq(John cleaned the room;Peter left) Peter left after John cleaned the room = seq(John cleaned the room;Peter left)
src	initial state, place, origin or source	The initial state, place, origin or source of an entity or event.	John came from NY = src(came;NY) John is from NY = src(John;NY) train from NY = src(train;NY) John changed from poor into rich = src(changed;poor) John received the book from Peter = src(received;Peter) John withdrew the money from the cashier = src(withdrew;cashier)
tim	time	The temporal placement of an entity or event.	The whistle will sound at noon = tim(sound;noon) John came yesterday = tim(came;yesterday)
tmf	initial time	The initial time of an entity or event.	John worked since early = tmf(worked;early)
tmt	final time	The final time of an entity or event.	John worked until late = tmt(worked;late)
via	intermediate state or place	The intermediate place or state of an entity or event.	John went from NY to Geneva through Paris = via(went;Paris) The baby crawled across the room = via(crawled;across the room)

Semantic Frames

Universal relations express generalizations on the types of lexical functions that are established between the UW and its arguments in the predication. These functions derive from the internal semantic structure of the UWs and are normally grouped into semantic frames, which describe the necessary semantic arguments that some UWs may have.

Semantic Frames of Verbs

Type	Frame	Description	Example
action verbs	agt	intransitive verbs whose subject is the agent of the event (unergative)	John danced = agt(danced;John)
	agt,obj	transitive verbs whose subject is the agent and the object is the patient of the event	John killed Peter = agt(kill;John)obj(kill;Peter)
	agt,cnt	transitive verbs whose subject is the agent and the object is the theme of the event	John read a book = agt(read;John)cnt(read;book) John talked about Peter = agt(talk;John)cnt(talk;Peter)
	agt,exp	transitive verbs whose subject is the agent and the object is an experiencer of the event	John congratulates Peter = agt(congratulate;John)exp(congratulate;Peter)
	agt,res	transitive verbs whose subject is the agent and the object is a result of the event	John made a cake = agt(make;John)res(make;cake)
	agt,plc	transitive verbs whose subject is the agent and the object is a place	John is going to NY = gol(go;NY) John is coming from NY = src(come;NY) John is going through NY = via(go;NY) John is going from NY to Geneva through Paris = src(go;NY)src(go;Geneva)via(go;Paris)
	agt,cnt,gol	ditransitive verbs whose subject is the agent and the objects are theme, and final state, place or recipient	John gave a book to Mary = agt(give;John)cnt(give;book)gol(give;Mary) John threw the ball at me = agt(throw;John)cnt(throw;ball)gol(throw;I)
	agt,cnt,src	ditransitive verbs whose subject is the agent and objects are theme, and initial state, place or origin	John took a book from Mary = agt(take;John)cnt(take;book)src(take;Mary)
	agt,exp,cnt	ditransitive verbs whose subject is the agent objects are experiencer and content	John told Peter about Mary = agt(tell;John)exp(tell;Peter)cnt(tell;Mary)
process verbs	obj	intransitive verbs whose subject is the patient (unaccusative)	John died = obj(die;John)
	gol,cnt	transitive verbs whose subject is the recipient (benefactive)	John won the lottery = gol(win;John)cnt(win;lottery)
experiential verbs	exp,cnt	transitive verbs whose subject is an experiencer	John likes Peter = exp(like;John)cnt(like;Peter)
	cnt,exp	transitive verbs whose object is an experiencer	Peter pleases John = cnt(please;Peter)exp(please;John)
stative verbs	aoj	linking verbs	John is dead = aoj(dead;John)
	plc	locative linking verbs	John is in NY = plc(John;NY) John is from NY = src(John;NY)
	aoj,cnt	possession verbs	John has two daughters = aoj(have;John)cnt(have;daughter)
	aoj,plc	locative verbs	John lives in NY = aoj(live;John)plc(live;NY)