Semantics Flashcards

1
Q

Productivity

A

can always generate new utterances by combining familiar parts of utterances in novel ways

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Components of Grammatical Theories

A
  • statements about what grammatical information is stored
  • statements about how you can recursively combine stored elements (and predict the properties of the bigger units that result from this process)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Principle of Compositionality

A

the meanings of complex expressions are determined by -

  • the meanings of simpler expressions they contain
  • the way the simpler expressions are combined
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Lexical Meaning

A

meaning = some sort of mental object (concept)

not always a strict relationship between linguistic meaning and the real world:
- London can refer to a place, the people that live there, the air above it, the buildings etc… (Chomsky)

relates to mental constructs which assign abstract properties even to things like London.

lexical meaning is part of an individuals implicit knowledge about language.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Concepts

A
concepts are not mental representations. otherwise, which dog would our concept of dog represent for example? our notion of dog must be a kind of 'essence', broad enough to cover:
> breed variation
> size and colour variation
> fictional/non-fictional
> talking/non-talking

cannot be captured with necessary and sufficient conditions:

  1. A dog is a mammal…
  2. …with four legs…
  3. …that barks…
  4. …that has fur.

Not all dogs have four legs, bark and have fur. The above only picks out most dogs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Prototypes

A

The best representation of a specific concept.
Best fruit? Apple or fig?
Best bird? Robin or ostrich?
(Eleanor Rosch)

Subjects are most likely to say the best bird is a robin because they are quicker to judge ‘a robin is a bird’ is true.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Relational Concepts

A

concepts are defined in reference to each other; part of the meaning of ‘cup’ is ‘not a bowl’.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Reasoning

A

we can reason most productively with concepts because we know certain relations hold between them:

  • hyperonym
  • hyponym
  • meronym
  • antonym
  • synonym
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Hyperonym

A

Refers to the category in which something belongs to.

Animal is a hyperonym of dog.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Hyponym

A

Refers to a member of a category.
Dog is a hyponym of dog.
Broader.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Meronym

A

Refers to an element of a concept.
Nose is a meronym of dog.
More general.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Antonym

A

Something that has an opposite meaning to something else.

Wet and dry are antonyms.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Synonym

A

Something that has the same meaning as something else.

Every groundhog is a woodchuck. They are synonyms.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Lexical Decomposition

A

The process of splitting complex concepts into simpler primitives.

If X kills Y, then X causes it to become the case that Y is not alive.

X kill Y cause (C, BECOME(not(alive(y))))

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Entailment

A

The truth of one element requires the truth of the other. If X killed Y then Y must be dead.

However, this is one-sided. If Y is dead, it does not necessarily follow that X killed Y.

Similar to hyponymy.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Hierarchal Relations

A

hyponymy and hyperonymy = transitive

If X is a hyponym of Y, and Y is a hyponym of Z, then X is a hyponym of Z also.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Inductive Reasoning

A

Some properties of being a bird are a matter of definition (having wings)
Other properties tell us what is normal (flying)

Through inductive reasoning, we can guess that kestrels can fly. Because:
kestrel is a hyperonym of hawk, a bird that flies.
and hawk is a hyponym of bird, an animal with wings.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Inductive Richness

A

Some distinctions are inductively richer than others (they tell us more about how they behave).
As a subset of things, ‘natural things’ does not share that many properties.
As a subset of animals, birds share quite a few properties: wings, beak, feathers, song…
Therefore, bird is inductively rich as a distinction.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Individuals and Properties

A

Individuals: Obama, London, a chair you’re sitting on…
Properties: being an ex-president, being a chair, liking sushi…

Properties of Obama:

  • being called Barack Obama
  • being American
  • being dignified (based on one’s own concept)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Predicates

A

semantic category
examples include: chair, happy, smoke
- denote the properties of being a chair, being happy, smoking…

distinct from any syntactic category - the above examples include a noun, adjective and verb.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Mental Models

A

consists of:

  • the individuals one is aware of
  • the properties one believe an individual has
  • the relations one believes an individual has
  • the relations once believes an individual stands in (loving, eating, injuring…)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Theory of Mind

A

the ability to recognise that other individuals have beliefs, that may differ from one’s own

  • develops gradually in childhood
  • impaired in some (some people with autism)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Intensions and Extensions

A

Extension - something that exists in the real world
Intension - one’s own concept of that thing

For example:
Extension: dogs in the real world
Intension: my own concept of doghood

Fictional objects, like unicorns, have intensions but no extensions (because they do not exist in the real world)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Intensional and Extensional Predicates

A

Certain predicates are extensional in the sense that they require us to interpret objects in the real world.
- Donald Trump discussed Obama
Others are intensional in that they concern hypothetical situations:
- Donald Trump kissed Obama

25
Q

Substitution Under Identity

A

Extensions shift with contexts; intensions don’t.
- Right now, Trump and POTUS have the same extension (but different intensions)

beliefs + situations + time = model of context
(provisionally) intension + context ~= extension

26
Q

Satisfaction Conditions

A

Understanding a request/question ~= knowing what would count as complying with that request/as an answer to the question

27
Q

Argument

A

The subject/individual combined with a predicate

28
Q

Truth Conditions

A

Understanding a statement ~= knowing the truth conditions of the statement.

Finding the truth conditions:
- “Barack Obama is very happy” is true IF the individual denoted by Barack Obama is in the set of very happy things (at the time in question)

29
Q

Paraphrase

A

One thing is true if and only if the other is:

I baked a cake.
A cake was baked by me.

Similar to synonymy.

30
Q

Contradiction

A

If one thing is true, the other must be false.

I baked a cake.
I have never cooked anything.

31
Q

Participant Roles

A

thematic relations

32
Q

Agent

A

animate individual that intentionally initiates an event.

JEREMY sang.

33
Q

Cause

A

initiator may not be acting intentionally but brings about a result.

THE WIND blew the roof off.

34
Q

Patient

A

an affected object.

Hilary ate AN APPLE.

35
Q

Instrument

A

a means with which an agent carries out an event.

THE KEY opened the lock.

36
Q

Theme

A

an object whose location, status, etc. may change.

I sent THIS PACKAGE to you.

37
Q

Source

A

the initial location of the theme.

The plane flew from AMSTERDAM to Houston.

38
Q

Goal

A

the final location of the theme.

The plane flew from Amsterdam to HOUSTON.

39
Q

Location

A

place where an event occurs.

He fell asleep IN THE BATHROOM.

40
Q

Benefactive

A

someone for whose sake an event is initiated.

Bill baked a cake for ROSE.

41
Q

Comitative

A

something in the presence of which an event occurs.

Jane went to church with BEN.

42
Q

Experiencer

A

an individual in whom an object or event induces some mental state

He fears MICE.

43
Q

1-Place Predicate

A

When there is only one open argument slot, meaning that is where the subject must go.

snore: EVENT, PRED = snore, THEME = X
John: j
John snores: EVENT, PRED = snore, THEME = j

44
Q

Unsaturated Predicate

A

A predicate that has an empty argument slot (snore)

45
Q

Saturated Predicate

A

A predicate with no empty argument slots (John snores)

46
Q

2-Place Predicate

A

When there are two open argument slots. We must make sure we associate the right argument with the right participant role.

fear: STATE, PRED = fear, EXPERIENCER = X, THEME = y
John fears Mary: STATE, PRED = fear, EXPERIENCER = j, THEME = m
Mary fears John: STATE, PRED = fear, EXPERIENCER = m, THEME = j

47
Q

Mapping Rules

A

Agents/cases are always subjects (exceptions being passiveS)

Patients are always objects (ditto)

Themes are objects unless there is no agent/cause/etc

48
Q

Hierarchy of Grammatical Function + Participant Roles

A

Subject > Direct Object > Indirect Object > Oblique

Agent/cause > Experiencer > Theme > Patient > Goal, location, source, etc

the higher on one hierarchy, the higher on the other)

49
Q

Idioms

A

Multiword expressions whose meaning cannot be determined on the basis of meanings of parts and the way they’re combined.

“kick the bucket”

  • big verb meaning ‘die’?
  • ‘kick’ has a special meaning when combined with ‘the bucket’?
50
Q

Passives

A

Patient realised as the subject instead of the agent (which is instead expressed as a modifier)

“The doctor cured the patient” becomes “The patient was cured by the doctor”

51
Q

Argument Structures for Active and Passive Verbs

A

Eat: EVENT, PRED=eat, PATIENT=x, AGENT=y
(includes an agent)

Eaten: EVENT, PRED=eat, PATIENT=x

Implies an active lexicon - operations can derive one lexical item from another

52
Q

Causatives

A

“The chef cooled the soup”
(Subject causes an effect on the object)

CAUSE(c, BECOME(STATE, PRED=cool, THEME=s))

53
Q

Inchoative

A

“The soup cooled”
(The effect is described without the agent being mentioned)
(However, it does not specify whether an agent is responsible for the cooling)

BECOME(STATE, PRED = cool, THEME = s)

The relationship between the causative and the inchoative is like the relationship between the active and the passive: if the soup was cooled by the chef, then the soup cooled (and was cooled)

54
Q

Stative

A

“The soup is cool”
(describes the end result of the inchoative sentence)

STATE, PRED = cool, THEME = s

55
Q

Propositional Attitudes

A

indicates that an individual can stand in relation to a proposition

believe: STATE, PRED = believe, EXPERIENCER = x, CONTENT = p
They believe that the Earth is flat = STATE, PRED = believe, EXPERIENCER = ‘they’, CONTENT = [STATE, PRED = flat, THEME =e]

can include recursion

  • I doubt they believe…
  • She said she doubts they believe…
56
Q

Recursive Compositional Semantic Theory

A

A predicate can take a proposition as an argument, forming a bigger proposition. Can be applied as many times as you want.

57
Q

Presupposed Propositions

A

Certain propositional predicates show the characteristic behaviour of presuppositions: invariance under negation

  • She knew that my cat was black.
  • She didn’t know that my cat was black

know P ~= believe P + presupposition that P is true.

58
Q

Control Predicates

A

Some verbs take non-finite clauses as complements

  • They persuaded him to leave.
  • She promised them to be there on time.

Different interpretations of ‘to-VP’

  • They persuaded him that he should leave.
  • She promised them that she would be there on time.

This implies that ‘to-VP’ can be interpreted as a proposition. - CONTROL.

59
Q

Componential Analysis

A

Represents word’s intension by breaking down into smaller semantic components

Man: human + male + adult