Outline of Semantics
Forms of thought
Mapping meaning onto language
Word meaning
Semantic features
Prototypes and Stereotypes
Relational meanings
(Word meaning and) longer expressions
Reference and Sense
Sentence meaning
propositions
sentence v.s. utterance
Discourse meaning : cohesion, coherence,
background knowledge, the cooperative principle
Markedness
Yun-Pi Yuan
1
Forms of thought
• A thought may be compared to a cloud
shedding a shower of words.
• Mental representation:
Have you ever had the experience of
wanting to express a thought, but you
couldn't find the words for it?
• Language is NOT the basic form of
thought.
Yun-Pi Yuan
2
Mental Representation
• Mental imageries:
A. sound images
B. visual images
C. math
D. movement—action patterns
Yun-Pi Yuan
3
Sound Images
• You can “play” music in your head, no?
• Reading music
Yun-Pi Yuan
4
Visual Images
• “Pictures in your mind”
• How do you find your way home?
• Remembering scenery: the apt. I stayed in
NYC, in Hsintien, and the one I stay now.
• Recognizing people: matching pictures
already in memory with what you see now.
• Painters: Michelangelo
• Matching color; dressing.
Yun-Pi Yuan
5
Math
• Doing math problems in your head.
(“Hsin Swan” 心算)
Yun-Pi Yuan
6
Movement (Action Patterns)
• How to tie knots; use tools; dance; write
Chinese calligraphy; tie your shoes; braid
hair; use chopsticks, etc.
• Books explained with pictures and words:
often easier just to follow pictures
Yun-Pi Yuan
7
Transfer among Different Forms of Thought
• Yes, we do it all the time:
e.g. We describe pictures in mind in words;
form pictures from words heard; put some
sort of process into math—then explain in
words; for dance draw pictures of steps,
etc.
• Therefore, language is not the basic form
of thought.
Yun-Pi Yuan
8
Semantics
• Definition: the study of meaning in
language; how meaning is
represented in language
• Importance of meaning: the basic function
of language is communication
• Difficulty to define “semantics” completely
Yun-Pi Yuan
9
Mapping Meaning onto language (1)
• Examples
English, Chinese, & Spanish: He gave me a pen.
Turkish: Babam bana topu verdi.
(father to-me ball gave)
(Nash 92)
actor recipient object action
(possessed by speaker)
(definite) (past, 3rd person, singular)
(witnessed by speaker)
Hebrew: Aba natan li
(daddy gave me
et
ha kadur.
the ball)
actor action recipient object definite object
particle
(past, 3rd person, singular, masculine)
Yun-Pi Yuan
10
Mapping meaning onto language (2)
• Problems of mapping meaning onto
language: not all straightforward beyond
the basic 4 elements (i.e., number, gender,
definiteness of participants and
where/how this information is to be
encoded).
Yun-Pi Yuan
11
Mapping meaning onto language (3)
• Examples of “my”: present different kinds
of possessions.
A. my shoes—can be thrown away when
worn out, but other
people
not likely to wear
them
B. my chair—but others can sit in it
C. my nose—has nothing to do with
others, nor will I “throw it
away”
Yun-Pi Yuan
12
Mapping meaning onto language (4)
• Note the differences:
A. He has a big nose.
(“Have”: I “possess” something; more general
than “own”)
B. *He owns a big nose.
(You cannot own parts of your body; only
materials or object which you can give away
or buy/sell it, can be owned.)
C. He is the possessor of his big nose.
(“Possess”—closer to “own” than to “have”)
Yun-Pi Yuan
13
Mapping meaning onto language (5)
• How does a child learn semantics?
• Slobin Model (Nash 91)
KNOWLEDGE of the world
Parts of KNOWLEDGE marked in HUMAN LANGUAGE
Parts of KNOWLEDGE marked in language X
(language the child is learning)
Yun-Pi Yuan
14
Mapping meaning onto language (6)
• The child’s problem of mapping meaning onto
•
language:
A. Which aspects of knowledge of world would
likely to be marked?
B. Which aspects must be marked in a
particular language?
C. How are they marked? (words; word
order, affixes, function word, …)
So, we’ll work hard to explain how some aspects
of our knowledge of meaning are expressed in
language.
Yun-Pi Yuan
15
Word meaning
• Word meaning including:
A. features
B. prototypes
C. stereotypes
D. relational meanings (degree, direction)
E. reference and sense (take us into semantics
of longer expressions)
Yun-Pi Yuan
16
Features
• Definition: more basic concepts/ideas that cannot be
•
“defined” any further; primitive semantic elements.
Combinations of features: [+ -] (e.g., see Nash 94-95)
A. Advantages
1. a universal element found in all langs. (Nash 95)
2. similar to phonological features
B. Disadvantage: very limited application
Yun-Pi Yuan
17
Advantage 1: Universal
• While we may speak different languages,
we’re all humans with the same human
brain, & perceive the world with the same
human senses.
e.g. [+HUMAN], [+ANIMATE], [+ROUND],
[+MALE], [+FEMALE], [+LIQUID],
[- MOVEABLE], etc.
Yun-Pi Yuan
18
Advantage 2: Similar to Phonological Features
• Psychologically similar to phonological
features
• Same kind of mental operation; from
phonology  semantics
• Phonemes: defined by its features
e.g. /p/=+consonantal, -voiced, +stop,
+bilabial
Yun-Pi Yuan
19
Disadvantage
• Very limited application—do not work for many
•
words
e.g. A. chair/stool/bench/bean bag
B. ugly/beautiful
C. red/green
D. table/desk
E. book/pamphlet
Lead to idea of prototypes
Yun-Pi Yuan
20
Prototype
• Definition: a typical/ideal example (serving to
•
represent the whole class); an examplar
Concept of prototype: helps explain meaning of
certain words in terms of resemblance to the clearest
examplar.
• Eleanor Rosch’s experiments:
A. bird:
Robin, sparrow, canary, dove, lark, parrot, owl, peacock,
duck, penguin, ostrich, bat
B. clothing:
shirts, dresses, skirts, bathing suit, pajamas, shoes,
stockings, the hat, gloves
C. vegetable:
pea, carrot, cauliflower, onion, potato, mushroom
Yun-Pi Yuan
21
Stereotype
• Definition: a list of typical characteristics of
describing something; more abstract
representation of possible qualities
e.g. bird: feathers, wings, beak, fly, lay eggs
Yun-Pi Yuan
22
Relational Meanings
• Words may differ +- a feature. But, many sets of
•
•
•
words differ, or may be grouped, in other ways,
including “degree” and “direction.”
Degree: amount—contrast to +- of features
e.g. hot/cold, long/short, tall/short, hard/soft,
good/bad, wet/dry, beautiful/ugly
Direction: for example, buy/sell, come/go,
give/receive, borrow/lend, read/write.
Note: A. “father”—also relational (in a different way)
B. kill and hurt—cause and effect relations
(Nash 97)
Yun-Pi Yuan
23
Longer Expressions
• Reference and sense: applying to semantics of both
•
words and longer expressions
Reference: dealing with the relationships between
language and the world (Nash 98)
e.g. “My son is in the beech tree.”
identify person
identify thing
• Sense: dealing with relationships inside the
language.
e.g. The moon was bright last night. (reference)
My love is like the moon. (sense)
Yun-Pi Yuan
24
Reference and Sense (1)
• Sense but not reference: function words, such as
•
and, or, never, perhaps, otherwise, but. These
make connections between meanings of different
units of language.
Same reference but different sense:
e.g. The evening star
west. (sunset)
Venus
The morning star
east. (sunrise)
Same object (same reference) but different sense
(different aspect); different ways of referring to
the same thing.
Yun-Pi Yuan
25
Reference and Sense (2)
• The same word can have more than one sense.
For example,“bank”
a. I have an account at the Bank of Scotland.
b. We steered the safe to the other bank of the river.
c. The DC-10 banked sharply to avoid a crash.
d. I banked the furnace up with coke last night.
e. a bank shot
Yun-Pi Yuan
26
Reference and Sense (3)
• Other examples:
my father/ the man who married my
mother/我先生/孩子的爸
different senses, although refer to the
same person (=same reference)
• Could have different reference
e.g. stepfather or illegitimate child
Yun-Pi Yuan
27
Sentence Meaning (1)
• Proposition= the basic idea/thought of the sentence;
•
events or states; say something about events/states.
Proposition: predicate +argument(s) (Nash 19-20. 84+)
Aspect of entity,
quality, state,
activity, relation with
other entity/ things.
entity (some
sort of thing)
• A sentence can have more than one propositions.
• sentence: (see definition)
utterance: what speakers say or write
Yun-Pi Yuan
28
Sentence
• Definition: a unit of language (an abstract
thing, a part of language itself), a string of
words put together by the grammatical
rules of a language.
• Meanings of a sentence come from only
within the language independent of
context.
Yun-Pi Yuan
29
Utterance
• Definition: the meaning of an utterance comes from both the
language & the context & from features of language (e.g.
intonation, stress, gestures)
• What speakers say or write: you can give the time, date,
place of an utterance (including intonation, stress, patterns
and gestures)
• An event in the world which can be thought as an example of
a sentence, or of part of a sentence (e.g., a phrase or a word)
• Different functions in context:
statement of fact
explanation
suggestion
denial
thanks
apology
tease
promise
insult
request, compliment
e.g. Mr. Nash likes tea. (Nash 20, 99)
argument
argument
predicate (shows relationship)
Yun-Pi Yuan
30
Sentence Meaning (2)
• Propositional meaning (sentence) vs.
interpersonal meaning (utterance)
• Proposition vs. utterance analysis
the difference: (Nash 100-101)
e.g. “The book is open.”—accusation
“Tom opened the book”— defense against
accusation; put
blame on
somebody else
Yun-Pi Yuan
31
Sentence Meaning (3)
• Examples of utterance:
•
“Can you open the window?”—mother to child
(order)
“Is your homework ready?”
–student
student (=can I copy it?)
–teacher
students (=now, turn it in)
Meaning of utterances based on the context
(depending on the interactions of the speakers
and their relationship).
Yun-Pi Yuan
32
Sentence Meaning (4)
• Sentence vs. utterance
e.g. He loves her.—sentence
“He loves her.”—utterance
(understand, but
who are they?)
(with knowledge
of reference of
pronouns)
• Expressions without propositional meaning, only
international meaning: e.g. Hello, Goodbye,
pardon, Hey (something like verbal gestures)
Yun-Pi Yuan
33
Sentence Meaning (5)
• Utterance meaning has to be determined
from the context (intentions of
speaker/hearer, their relationship; the time,
place, roles)
• Sentence meaning (propositions)
independent of context.
Yun-Pi Yuan
34
Sentence Meaning (6)
• Practice:
Utterance Sentence Propositions
Can be loud or quiet
Can be grammatical
or not
Can be true or false
In a particular
regional accent
In a particular
language
Yun-Pi Yuan
35
Sentence Meaning (7)
• Practice:
Utterance Sentences propositions
Can be loud or quiet
+
-
-
Can be grammatical
or not
Can be true or false
+
+
-
+
+
+
In a particular
regional accent
In a particular
language
+
-
-
+
+
-
Yun-Pi Yuan
36
• Family tree relationship:
proposition
sentence
sentence
sentence
utterance utterance utterance utterance utterance utterance
Yun-Pi Yuan
37
physical actions
[gesture]
mental processes
[thoughts]
abstract semantic entities
[propositions]
Linguistic entities
[e.g. sentences]
Actions [e.g. utterances]
Yun-Pi Yuan
38
Discourse (1)
• Language longer than a sentence
• Important at many levels: syntax+
morphology; discourse structures—the
structures of units longer than a sentence.
• TEXT e.g.: (Nash 101)
The monster danced with Yang Li-Hua. He
enjoyed it. She didn’t.
Yun-Pi Yuan
39
Discourse (2)
• Examples of different discourse structures
A. writing
a. paragraph
b. composition (longer organization)
c. book (chapter…)
d. story—typical structure: chronological
order
B. apartment descriptions
C. conversation: casual/classroom/ ordered
discussion/debate/interview/ritual (e.g.
church ritual, graduation, wedding ritual,
classroom ritual—起立.敬禮.坐下.報告)
Yun-Pi Yuan
40
Discourse (3)
• Some important elements in discourse:
cohesion, coherence, background
knowledge, the co-operative principle
• Cohesion: “the ties and connections which
exist within texts.” Text: a piece of spoken
or written language.
Yun-Pi Yuan
41
Cohesion (1)
• Examples of cohesion: (Yule 140)
pronouns, (e.g. he, my, I , it); lexical
connections (e.g. Lincoln convertible—the
car—the convertible); general connections
with shared meaning elements (e.g.
“money”—bought—sawing—penny—
worth a fortune—sold—pay); relationship
marker (e.g. “however”); tense—first 4
sentences: past tense, last one: present—
a different time.
Yun-Pi Yuan
42
Cohesion (2)
• Cohesion: the grammatical and/ or lexical
relationships between the different
elements of a text. This may be the
relationship between different sentences
or between different parts of a sentence.
Example:
A: Is Jane coming to the party?
B. Yes, she is.
There is a link between Jane and she, also
between is… coming and is.
Yun-Pi Yuan
43
Coherence (1)
• The relationships which link the meanings of
•
utterances in a discourse or of the the sentences
in a text.
These links may be based on the speakers’
shared knowledge (background knowledge)
e.g. A: Could you give me a ride home?
B: Sorry, I’m visiting my sister.
There’s no grammatical or lexical link between
A’s Q and B’s reply, but the exchange has
coherence, because both A and B know that B’s
sister lives in the opposite direction to A’s home.
Yun-Pi Yuan
44
Coherence (2)
• Coherence: that the text makes sense—
coherence achieved more by people than
by texts (than by language itself)—we
expect coherence—we “try to arrive at an
interpretation which is in line with [our]
experience of the way the world is” (Yule
141).
• An example of coherence without
cohesion
(Yule 142)
Yun-Pi Yuan
45
Coherence (3)
• Obviously, there’s something else involved
[what is it?] in the interpretation of a
conversation, except the information
expressed in the sentences.
• It is clear that language users must have a
lot of knowledge of how conversational
interaction works which is not simply
“linguistic knowledge.”
Yun-Pi Yuan
46
Background knowledge
• Examples (Yule 146-47)—inference, build-up,
•
changing influence
2nd sentence: Who is John?
How’s he traveling?
3rd sentence : Who’s John? (How traveling?)
4th sentence: Who’s John?
5th sentence: surprise
We create what the text is about (not just the
text does this), based on expectation of what
normally happens (=background knowledge).
Yun-Pi Yuan
47
The Cooperative Principle (1)
• In conversation participants are assumed (by
•
others) to be cooperating.
Four Maxims: set out by Grice (1975)
Quantity: an informative as is required, no
more, no less.
Quality: Don’t say something you believe
to be false or something you don’t
know.
Relation: Be relevant
Manner: Be clear, brief and orderly
Yun-Pi Yuan
48
The Cooperative Principle (2)
• These are the normal expectations
e.g. expectations about Quantity: “To
make a long story short,” “I won’t bore
you with all the details.”
Quality: “An far as I know;” “Correct me if
I am wrong;” “I think;” “I feel;” “It’s
possible that…” (“maybe”)
Yun-Pi Yuan
49
The Cooperative Principle (3)
• The 4 maxims and the whole principle 
allow interpretations (see Yule 145 bottom)
Carol: Are you coming to the party tonight?
Lara: I’ve got an exam tomorrow.
• Just a brief introduction to Discourse—many
more elements involved, very complex.
Yun-Pi Yuan
50
Lexicon (1)
• Q: Do the lexical items (words) of a
language have some sort of overall
structure/organization like phonology,
morphology, and syntax have?
• What’s the exact nature of a unit for
definition? That is, what is a lexical unit?
Yun-Pi Yuan
51
Lexicon (2)
• Dictionary entry is not exactly what we think of
as a word. It’s really a paradigm: an example of
all the forms of a word, used to represent the
whole set.
Examples:
child—represents child, child’s, children,
children’s
take—takes, taking, took, taken.
Some sets include only one member: how, yet,
often
Yun-Pi Yuan
52
Lexicon (3)
• How is the paradigmatic form chosen?
e.g. find a new word
ritualistic  look up ritual
larger look up large
Yun-Pi Yuan
53
Unmarked
• The paradigmatic form is the unmarked form:
the form which does not seem “special” in any
way, that seems most “basic”, that has nothing
added (phonemes, sounds, morphemes).
e.g. child: child’s, children
large: larger
car: cars
ritual: ritualistic
strangle: strangulation
old/young: “How old is she?” the normal Q
Yun-Pi Yuan
54
Markedness (1)
• The theory that in the languages of world certain
•
linguistics elements are more BASIC, National, and
Frequent (these elements are unmarked; less basic,
national, frequent elements are marked)
Examples:
A. Singular examples: car—cars
(singular=unmarked; plural=marked)
B. unmarked: S-V-O sentence: I dislike such people.
marked: O-S-V sentence: Such people I dislike.
Yun-Pi Yuan
55
Markedness (2)
• Marking may be a basic principle for
assigning universal (and possibly innate)
values to certain kinds of features
Slobin Model
(Nash 91)
Yun-Pi Yuan
56
Markedness (3)
C. more frequent=unmarked
e.g. Falling intonation vs. rising intonation
D. more specific=marked
(more common=unmarked)
e.g. dog (unmarked) vs. bitch (marked)
E. distribution—unrestricted (unmarked)
e.g. How tall is John? vs. How short is John?
Yun-Pi Yuan
57
Markedness (4)
• Markedness theory applies at all levels:
A. Phonology
e.g. /p, t, k, s, n/ unmarked consonants
/v, z, Q, ð/ more marked
falling intonation=unmarked
rising intonation=marked
B. lexicon e.g. dog/ bitch (marked)
C. morphology e.g. car/ cars (marked)
D. syntax e.g. active vs. passive (marked)
Yun-Pi Yuan
58
Markedness (5)
• Discourse e.g. politeness
too polite
(marked)
Would you be so
kind? just let me
borrow your
pencil for a
minute?
unmarked
Could you lend
me a pencil?
Yun-Pi Yuan
too informal
(marked)
Without saying
anything, just
grasp the pencil.
59
Markedness (6)
• Unmarked elements: easier to acquire
• Marked elements: more difficult to acquire
• Some experimental evidence shows that
teaching marked forms can lead to faster
acquisition of both marked and unmarked
forms, but teaching unmarked forms won’t
help students learn marked forms.
Yun-Pi Yuan
60