Term
| This is why a METALANGUAGE might be useful. |
|
Definition
Metalanguage is good for clarifying scope ambiguity:
"Mary thought she read every book on the list"
-could mean she knew abuot list and was pretty sure she'd read all those books, OR
-could mean that everytime she was asked about a book from the list she'd never seen, she was pretty sure she'd read it. |
|
|
Term
| In a nutshell, this is what Frege believed about compositionality. |
|
Definition
| Frege believed that both sense and reference are necessarily compositional. |
|
|
Term
| This is what Austin (1955, 1961) thought of speech acts (think performatives). |
|
Definition
Austin argued that even declarative sentences (which he called constatives) had the performative function of 'stating'.
Thus, all utterances are speech acts of some kind. |
|
|
Term
| Cross-linguistically, all langauges seem to have these three types of sentences. |
|
Definition
| declarative, interrogative and imperative |
|
|
Term
| According to Austin's Speech Act Theory (1961), we might say that there are two basic parts to a sentence's meaning. |
|
Definition
-its conventional meaning (proposition)
-the speaker's intended speech act |
|
|
Term
| Generally speaking, what do the words REFER/DENOTE mean? |
|
Definition
Words refer/denote items in the world by picking them out or identifying them.
That entity is the referent(reference)/denotation.
denote: relationship between word and world.
refer: relationship between mind and world (the act of the speaker picking entities out for denotation)
Thus, denotation is a stable, enduring relationship, while reference changes depending on context (different people might apply the same word to different things under different circumstances). |
|
|
Term
There are two general models for describing how meaning is conveyed through words:
1. The Referential (denotational) approach
2. The Representational approach |
|
Definition
1. The Referential(denotational) approach holds that words and senteces have meaning because they denote real-world items and situations. (word-world)
2. The representational approach holds that a language represents a theory about reality. Thus (the same) reports about the world are influenced by (different) mental models and conceptual structures conventionalized in languages.--Meaning arises from language reflecting our conceptual frameworks. (mind-world) |
|
|
Term
Note: In examining the Referential approach to meaning, we FIRST note that some words never refer to things in the world (like the words so, maybe, very, if...). These can be said to be Non-Referring expressions. SECOND, people sometimes use even Referring words such that they do not refer, as with generic use:
"A cholecystectomy is a serious procedure." (non-referring) |
|
Definition
THIRD, we also note that certain words and expressions always refer to the same thing (Eiffel Tower, Pacific Ocean...), whereas the reference of other words depend on context, like pornouns (I, you...). THIS IS THE DIFFERENCE BETWEEN constant AND variable reference.
Deictics: words whose denotational capability needs contextual support. |
|
|
Term
Names and NPs are paradigmatic examples of referring words.
Two competing theories have arisen to explain how NAMES achieve reference.
description theory
causal theory |
|
Definition
Bertrand Russel championed the DESCRIPTION THEORY: names are a label/shorthand for knowledge of a referent.
The CAUSAL THEORY: At the point when a name is given, people witness the use of the name and they perpetuate it. This is the grounding of the name. This process of propagation continues over time/space (as with famous people). In this way, people come to know and recognize the name of someone about whom they know very little or nothing at all.
(These theories about names can also be applied to natural kinds.) |
|
|
Term
NPs can be definite or indefinite, and applied distributively or collectively. (individual members vs. whole as a body)
These are the 3 main problems with reference as a theory of meaning... |
|
Definition
1. It predicts many words have no meaning (so, not, very, but, of) when clearly they do.
2. Many words refer to things that do exist/have not ever existed (unicorn, WWIII).
3. The same thing can be referred to with different expressions, as with the NAME and the DEFINITE DESCRIPTION (Anwar al Sadat, president of Egypt) |
|
|
Term
In terms of Reference as a theory of meaning, three problems arise:
Some words are predicted to have no meaning (so, if, and...)
Many words refer to things that don't exist (unicorns)
There are multiple expressions for a given referent (name, description)
These are some comments on the third issue... |
|
Definition
a. In these cases, we would probably say that even though the denotation is the same, the meaning differs.
b. We can even use different expressions to refer to the same referent but not know the referent is the same (morning star=evening star=Venus)
**This all suggests that there is more to meaning than reference. |
|
|
Term
| Mill (1843) made a similar distinction between denotation/connotation that Frege (1882) made about reference/sense. |
|
Definition
Denotation/reference: The set of all THINGS in the world of discourse that can properly be referred to in that way.
Connotation/sense: The set of PROPERTIES that things must have in order to be referred to in that way.
Frege defined 'sense' as the "mode of presentation" of a sign, which allows speakers to identify the referent; knowledge shared by all users of a language. |
|
|
Term
This is an explanation of Frege's (1882) sense vs. reference:
a. Obama is the first black president of the United States.
b. Obama is Obama. |
|
Definition
(b) is an obvious tautology. The fact that, while both expressions refer to the same individual, exchanging one for the other does not carry the same meaning suggests that 'meaning' involves more than just reference.
That is, even thogh they are co-referential, they carry different senses.
Thus, to Frege, "Bob Dylan is Robert Zimmerman" includes two expressions which have the same referent but two different senses. Therefore, he would say that this statement is not analytically true (and IS informative). |
|
|
Term
| This is Frege's theory of compositionality. |
|
Definition
Frege assumes that sentence meanings are compositional and predictable from teh meanings of words + the syntax by which they're combined.
He believes that both reference and sense must be compositional. |
|
|
Term
| This is what Frege has to say about the relationships between sense, reference, the Law of Substitutivity, and truth value. |
|
Definition
Frege demonstrates that the reference of a sentence cannot be its proposition. A PROPOSITION can be true or false--something people can believe.
The reference of a sentence cannot be its proposition due to the Law of Substitutivity. That is, it should be the case that if a=b and b=c, then a=c, BUT people can claim a fact about the morning star while not knowing that it also holds true for the evening star (Venus).
These sentences cannot have the same proposition because one can believe one but not the other. Thus, Frege concludes that the proposition (the SENSE) of the sentence is changing, while the reference (the TRUTH VALUE) is not changing.
So...
Proposition=sense of a sentence
truth value=reference |
|
|
Term
| Frege's claim that sense=proposition and reference=truth value also helps explain NPs with no referent, like "the fountain of youth". |
|
Definition
A sentence which contains an empty NP cannot have a truth value, since there is no reference. But it is not meaningless because it still expresses a proposition (a sense).
For Frege, no reference = no truth value = truth value gap. |
|
|
Term
| These are some examples of PROPOSITIONAL ATTITUDE VERBS. |
|
Definition
believe, hope, want, deny...
(they all take a complement clause; they all express the speaker's attitude or understanding about the proposition) |
|
|
Term
| This is what Frege says about the Law of Substitutivity and propositional attitude verbs. |
|
Definition
Frege pointed out that the Law of Stubstitutivity does not hold in the complements for propositional attitude verbs. This law can justify that (a) and (b) have the same (and not opposite) truth values, and why (b) and (c) have opposite (not the same) truth values:
a. Mary knows [that the morning star is Venus].
b. Mary doesn't know [that the evening star is Venus].
c.#Mary doesn't know [Venus is Venus]. |
|
|
Term
| This is how the Law of Substitutivity relates to REFERENTIAL OPACITY. |
|
Definition
The Law of Substitutivity doesn't apply here because these sentences are referentially opaque; the reference shifts.
a. T=Copernicus believed [T=that the earth revolves around the sun].
b. F=Copernicus believed [T=that the planetary orbits are ellipses].
(Complements are mentions, not uses, with regard to their truth value.) |
|
|
Term
a. Oedipus wants to marry Jocasta.
b. Jocasta is Oedipus' mother.
---
c. Therefore, Oedipus wants to marry his mother.
This reasoning is not valid because some NPs are referentially opaque and the sentence has a propositional attitude verb ('want'). SUBSTITUTION IN SUCH CONTEXTS DOES NOT PRESERVE TRUTH VALUE. |
|
Definition
There is a de dicto/de re ambivuity here. De dicto is (c). The de re interpretation is that the person Oedipus happens to wan to marry happens to be his mother.
Frege did not notice this ambiguity. Bertrand Russell did, and offered a solution (cf. definite NPs as quantifiers). |
|
|
Term
| Note: Russel rejected the distinction between sense and reference. Instead, he said that definite NPs are really QUANTIFIERS. And definite NPs, like QPs, without a predicate don't mean anything. |
|
Definition
|
|
Term
| Note: Semantics has a long history: often ancient Greeks and Sanskrit grammarians are cited in modern scholarship. |
|
Definition
|
|
Term
| This is the difference between semantics and pragmatics. |
|
Definition
Semantics (meaning): the study of inherent, linguistic meaning of words and sentences.
Pragmatics (context): the study of the meaning of utterances; thos aspects of meaning that depend on context and usage. |
|
|
Term
Ambiguity is a large topic of focus for semantics.
Here are examples of structural, lexical and aspectual ambiguity. |
|
Definition
the "synthetic buffalo hides" comic (structural)
"A boiled egg is hard to beat" (lexical)
"Baked since 1919" (aspectual)
Liars paradox: "This statement is a lie."--true or false? |
|
|
Term
| Note: To know the meaning of a declarative sentence is to know what the world would have to be like for the sentence to be true. |
|
Definition
|
|
Term
| Note: Semantic properties of a word are language-specific. You must specify the target language. |
|
Definition
| One big problem in semantics though is sorting out and keeping straight the difference between the target language and the metalanguage. |
|
|
Term
| Often, compositionality is appealed to in seeking the meaning of compound words and phrases. Why may this be a problem? |
|
Definition
This can be a problem because some words have grammatical and not lexical meaning, some have both, etc.
"You're daddy's rich AND your mamma's good looking."
Truth table for 'and':
P Q P and Q
T T T
T F F
F T F
F F F |
|
|
Term
| Another word that causes problems for compositionality is 'but'. |
|
Definition
-sometimes it signals a simple contrast
-sometimes a counter-expectation:
"My parents were poor, but honest."
[Poor][Honest][Poor but Honest]
T T T
T F F
F T F
F F F
These truth tables show that 'and' and 'but' are at least partially synonymous.
'But' has extra pragmatic meaning, but is semantically the same as 'and'.
**This linkage underscores the reason for studying semantics AND pragmatics together. |
|
|
Term
| Types of SEMANTIC knowledge... |
|
Definition
synonymy
contradiction
ambiguity
entailment |
|
|
Term
Semantics is historically related to (and grew out of) semiotics.
a. Both study human tendency toward signification.
b. Semantics is the study of linguistic meaning.
c. Semiotics is the general study of sign systems.
d. C.S. Peirce provided the distinctions of ICON, INDEX, SYMBOL... |
|
Definition
C.S. Peirce:
icon: similary between sign and what it represents
index: sign is closely relate to what it represents
symbol: only a conventional link between sign and signified |
|
|
Term
| Saeed (2009) lists three principle challenges to semantics (Kroeger lists two more). |
|
Definition
Circularity-you must use words to define other words (we rely on: metalanguage)
Encyclopedic vs. linguistic knowledge- knowing where one stops and the other starts (we rely on: the notion of a mental lexicon for a person's linguistic knowledge)
Context-oftencontributes to or even determines meaning, but how can we include that in definitions? (we rely on: an assumed split in meaning of an expression in/out of context--semantics (literal meaning) and pragmatics (the role of context and processes we use to figure out meaning)) |
|
|
Term
| These are three recognized levels of meaning. |
|
Definition
word
sentence
utterance (Saeed's 'speaker meaning') |
|
|
Term
| This is the difference between a sentence and an utterance. |
|
Definition
sentence: a well-formed group of words
utterance: a speech event by a particular speaker in a particular context (at a particular time). "the totality of what the speaker intends to convey by making an utterance" (Cruse 2000) |
|
|
Term
| These are three ASPECTS OF MEANING. |
|
Definition
1. what is said (semantic content)
2. what is intended (implicature)
3. what is done (SPEECH ACT)
"Do you want some popcorn?"
Literal meaning=request for info
Speech act=offer of popcorn |
|
|
Term
Note: In terms of semiotic categories, semantics concerns symbols since mos words are symbols.
(but onomatopoetic words are iconic) |
|
Definition
In terms of the triangle between
MIND
/ \
SIGN WORLD
Cognitive semantics is interested in the mind-sign connection.
Denotational semantics is interested in the sign-world connection. |
|
|
Term
| In terms of semantics, what information is stored in the mental lexicon (vs. simply not stored)? |
|
Definition
Words and derivations are in the mental lexicon, but sentences are not--they are infinite and unpredictable. We make new sentences all the time.
We rarely create neologisms (new words) though, and when we do hear one, we immediately know "it's not a real word." |
|
|
Term
| Generally speaking, this is COMPOSITIONALITY. |
|
Definition
Compositionality: the meaning of a sentence is predictable from the meanings of the words and rules combining them.
That hearers understand speakers indicates that semantic and pragmatic meaning are rule-governed.
e.g., a "yellow submarine" is found in the intersection of the sets of yellow and submarine (think Venn diagram). |
|
|
Term
| Saeed (2009) suggests three challenges to accounting for meaning; Kroeger adds go that and suggests 5 challenges. |
|
Definition
1. Circularity: need metalanguage to try to overcome this
2. Linguistic vs. Encyclopedic knowledge
3. Context-driven variation in meaning
4. Objectivity and precision (going beyond vague intuitions)
5. Analysis based on linguistic evidence |
|
|
Term
| These are five types of figurative language. |
|
Definition
Hyperbole
Euphemism
Metonymy
Synechdoche
Litotes |
|
|
Term
| This is what METAPHOR means. |
|
Definition
| an implicit comparison based on one or more shared properties. |
|
|
Term
| This is what EUPHEMISM means. |
|
Definition
using an inoffensive term to replace an offensive term ('powder my nose' vs. 'pee')
From Bible: 1 Sam 24:2-3, Saul went into the cave..."to cover his feet" (go to the bathroom)
"Drumstick" is a euphemism from when "leg" was considered inelegant. |
|
|
Term
| This is what LITOTES means. |
|
Definition
understatement; asserting a proposition by negating its opposite.
From Bible: Jer 30:19b, "I will multiply them and they SHALL NOT BE FEW. I will make them honored and they SHALL NOT BE SMALL." |
|
|
Term
| This is what METONYMY means. |
|
Definition
identifying a referent by something associated with it.
"Glossa" meaning both 'tongue' and 'language'
Old English "bead" came from the term meaning 'prayer'
From Bible: Acts 8:27-8, "Ad behold, and Ethiopian...was returning; seated in his chariot, he was READING THE PROPHET ISAIAH."
Matthew 3:5-6 "Then went out to him JERUSALEM AND ALL JUDEA AND ALL THE REGION ABOUT THE JORDAN." |
|
|
Term
| This is what SYNECDOCHE means. |
|
Definition
a form of reference where a part stands for a whole.
From Bible: John 6: 30-1, "...OUR FATHERS ate the manna in the wilderness..."
Acts 5:9 "...Hark, THE FEET of those that have buried your husband are at the door..." |
|
|
Term
| This is what HYPERBOLE means. |
|
Definition
exaggeration.
From Bible: John 3:31-2, "He bears witness...yet NO ONE receives his testimony..."
Mat. 3:5-6 "Then went out to him JERUSALEM AND ALL JUDEA AND ALL THE REGION ABOUT THE JORDAN." |
|
|
Term
| This is what METAPHOR means. |
|
Definition
an implicit comparison between to un-like things based on one or more shared properties.
"Pastor" is a metaphor.
From Bible: Acts 26:14 "Saul, Saul, why do you persecute me? It hurts you to KICK AGAINST THE GOADS." |
|
|
Term
| Here is one view of where figurative usage of a word comes from. |
|
Definition
| An interesting question to ask is WHERE the figurative use comes from. One explanation holds that figurative meanings are calculated in the minds of hearers, when confronted with non-literal language. Over time, the novel metaphors cement (fossilize) and become "dead metaphors" or clichés. |
|
|
Term
| The English word "pioneer" is an example of the development of both metaphor and synecdoche. |
|
Definition
"Pioneer" comes from Old French 'peonier' meaning 'foot soldier', which itself comes from Latin 'pedones' derived from 'pes/pedis' meaning 'foot'.
This development involves both metaphor and synecdoche. |
|
|
Term
Note: A semantic theory examining a native speaker's competence in English should reflect his knowledge of:
Synonymy
Entailment
Contradiction
Presupposition
Tautology |
|
Definition
|
|
Term
| These are basic definitions of ENTAILMENT and PRESUPPOSITION. |
|
Definition
Entailment- knowledge of one thing automatically leads to knowledge of another.
Presupposition- one thing is part of the assumed background against which another is said. |
|
|
Term
| These are Aristotle's LOGICAL ARGUMENT TYPES. |
|
Definition
Modus ponens
Modus tollens
Hypothetical syllogism
Disjunctive syllogism |
|
|
Term
| This is an example and truth table for MODUS PONENS (one of Aristotle's logical argument types). |
|
Definition
Modus ponens:
If Arnd left work early, then he is at the pub.
Arnd left work early.
---
He is at the pub.
p → q
p
---
q |
|
|
Term
| This is an example and truth table for MODUS TOLLENS (one of Aristotle's logical argument types). |
|
Definition
Modus tollens:
If Arnd has arrived, then he is at the pub.
Arnd is not at the pub.
---
He has not arrived.
p → q
~q
---
~p |
|
|
Term
| This is an example and truth table for HYPOTHETICAL SYLLOGISM (one of Aristotle's logical argument types). |
|
Definition
Hypothetical syllogism:
If Arnd is at the pub, then he is drinking beer.
If Arnd is drinking beer, then he is drinking Guinness.
---
If Arnd is in the pub, he is drinking Guinness.
p → q
q → r
---
p-->r |
|
|
Term
| This is an example and truth table for DISJUNCTIVE SYLLOGISM (one of Aristotle's logical argument types). |
|
Definition
Disjunctive syllogism:
Arnd is in the public bar or he is in the lounge.
Arnd isn't in the public bar.
---
Arnd is in the lounge.
p or q
~p
---
q
**The study of the truth effects of connectives like these four argument types is called PROPOSITIONAL LOGIC. |
|
|
Term
Note: Truth, truth tables...
Empirical (contingent) truth: we have to have access to the facts to assess whether the statement is true.
Semanticists call a sentence's being T or F its TRUTH VALUE, and the facts that must occur in order to render a sentence T its TRUTH CONDITIONS. |
|
Definition
The black and white (no middle ground) nature of truth tables may present problems for actual, natural language.
Logicians use a schema called LOGICAL FORM to demonstrate the truth values of statements. Lower case letters=statments and special symbols=linguistic operations that alter their values. (see basic truth tables for 'and' and 'or' and 'material implication') |
|
|
Term
Note: Some truths do not seem to depend on any real-world facts, but rather are true as a function of linguistic structure. For example, tautologies are always true and contradictions are always false.
Thus, we may contrast A PRIORI truth with A POSTERIORI truth. |
|
Definition
So too there is Leibnitz's distinctions between NECESSARY and CONTINGENT truths. Necessary truths cannot be contradicted irrespective of facts, whereas contingent truths are dependent on facts.
Statements whose truth value follows the meaning relations of the sentence, regardless of any relationship with the world, are ANALYTIC.
Those that become true only when they accord with known facts of the world are SYNTHETIC. |
|
|
Term
| Thus, there are 3 related truth distinctions (between truths that area always true, and truths that need context). |
|
Definition
A priori vs. a posteriori (epistemological distinction)
Necessary vs. contingent (metaphysical distinction)
Analytic vs. synthetic (semantic distinction) |
|
|
Term
| This is what PREDICATE LOGIC and LOGICAL WORDS are. |
|
Definition
Predicate logic- the study of the truth behavior of sentences with quantifiers like ALL, EVERY, EACH, SOME, ONE.
Logical words- words like connectors and quantifiers that can influence the truth value of statements. |
|
|
Term
| This is what an ENTAILMENT is. |
|
Definition
Entailment- a special semantic relation because it involves fixed truth relations between sentences that hold regardless of teh empirical truth of the sentences.
Sentence P entails sentence Q when the truth of the first P guarantees the truth of Q, AND the falsity of Q guarantees the falsity of P.
(see entailment truth table) |
|
|
Term
| In terms of entailments, this is how we might define PARAPHRASE. |
|
Definition
The relationship of entailment permits us to define PARAPHRASE as a sentence that has the same set of entailments as another.
(something similar could be said for synonymy) |
|
|
Term
| See composite truth tables for SYNONYMY and CONTRADITCTION. |
|
Definition
(with synonymy, both sides of the truth table must be true, or both must be false, regardless of which way the arrow goes.)
(with contradiction, the truth values must always be opposite, regardless of which direction the arrow goes.)
Thus, overall truth-based description seems to work well for most of the key forms of semantic knowledge a speaker is said to process...but PRESUPPOSITION is one real problem for this approach! |
|
|
Term
Note: Entailment
The relation of entailment (like analyticity) exists by virture of linguistic structure, NOT empirical facts.
The source of entailment may be lexical or syntactic (in that passive & active sentences entail each other) |
|
Definition
| Hyponomy is a regular source of lexical entailment (i.e., 'dog' entails 'animal') |
|
|
Term
| This is how Cruse (2000) defines ANALYTIC sentences. |
|
Definition
Cruse (2000) "...sentences which automatically express true propositions in any context, by virtue of the meanings of their constituent words and their arrangement."
(i.e., truth follows from linguistic structure, not knowledge about the world (Kroeger))
(AKA a priori; tautology; necessary) |
|
|
Term
| This is a how we might define SYNTHETIC sentences. |
|
Definition
opposite of analytic; you must know or find out something about the word in order to assess the truth value of the sentence.
(AKA a posteriori; contingent) |
|
|
Term
|
Definition
| Paradox: necessarily always false because the statement is self-contradictory. |
|
|
Term
| Note: Tehre is no formal logical symbol for entailment because it is not a logical relation--it's semantic. |
|
Definition
|
|
Term
| These are some of the things Frege (1892) contributed to semantics. |
|
Definition
Frege (1892)
1. IDentity statements
2. distinguishing sense from reference
3. sense/reference of whole sentences (think compositionality)
4. Propositional attitudes
5. Proper names |
|
|
Term
| This is an IDENTITY STATEMENT and Frege's position on them. |
|
Definition
Identity statement:
a. Bob is my neighbor (synthetic)
b. Bob is Bob (analytic)
Clearly, identity statements are not just about the referent. Using different senses (different expressions) to refer to the same referent makes identity statements INFORMATIVE, not just tautological. |
|
|
Term
| This is how Frege would define SENSE. |
|
Definition
| The sense of a phrase gives a "MODE OF PRESENTATION" of its referent. (1884) |
|
|
Term
| This is what Frege would say about SENTENCE-LEVEL sense and reference. |
|
Definition
sense=proposition (changes if co-referential expressions are used)
reference=truth value (is constant) |
|
|
Term
| This is what Frege had to say about PROPOSITIONAL ATTITUDE VERBS. |
|
Definition
Propositional attitude verbs show a relationship between sentient being and sentence complement (like quotes, kind of, because truth values don't hold). --There's not a relation between truth values of matrix and complement.
(cf. referential opacity / opaque contexts)
No Law of Substitutivity for quotes (mentions) and propositional attitude complements. |
|
|
Term
| This is what Frege had to say about proper names. |
|
Definition
Proper names are like definite descriptions--same reference and (tolerably) different senses.
(but this is problematic) |
|
|
Term
| These are some of the things Russell (1905) contributed to semantics. |
|
Definition
1. quantification
2. definite descriptions
3. propositional attitudes
4. proper names |
|
|
Term
| Here is an example of how Russel would treat indefinite articles. |
|
Definition
a. A dog is barking.
∃x[dog'(x) & is-barking'(x)]
('a' is a quantifier)
b. Rover is a good hunter.
good-hunter'(Rover)
('a' is just a grammatical necessity) |
|
|
Term
| This is what Russell says is the difference between definite and indefinite articles. |
|
Definition
'the' means ∃x and ∀x and y→x
("at least 1 and only 1") |
|
|
Term
| This is what Russell (1905) would have to say about scope. |
|
Definition
De dicto=narrow scope
Oedipus wanted
∃x∀x[[Oed's mom'(y)<-->y=x] & marry (Oed, x)]]
(OPAQUE; incest)
De re=wide scope
∃x∀x[[Oed's mom'(y)<-->y=x] & Oed wanted [marry(Oed, x)]]
(TRANSPARENT; a horrible mistake) |
|
|
Term
| This is what Russell (1905) had to say about proper names. |
|
Definition
| Mill (1843) said proper names ONLY denote; Russell and Mill agreed on a descriptional view of proper names--semantically equivalent to definite descriptions. |
|
|
Term
| Abbot (2010) gave a historical sketch of sense and reference, including descriptions of these big names in the field. |
|
Definition
Mill (1843)
Frege (1882)
Russel (1905)
Basic explanation/argumetn for the importance of compositionality in human language is that you can understand most things (not idioms) on first use. |
|
|
Term
| Cruse (2004) talked about this. |
|
Definition
Contextual variation of word meanings:
A single sense can have multiple FACETS.
Facets are distinguishable components of a global whole, but not capable of being subsumed under a hypernym. Fully discrete but non-antagonistic readings of a word (Cruse 2000)
examples: book, bank, Britain, duck/chicken... |
|
|
Term
| In terms of FACETS (think Cruse 2000, 2004), where did the idea first come from? |
|
Definition
| Nunberb (1979) called this concept "DENSE METONYMY" |
|
|
Term
| These are four pieces of evidence for the discreteness (autonomy) of facets (Cruse 2000. 2004). |
|
Definition
Evidence for discreteness (autonomy) of facets:
1. Distince collocational/selectional restrictions; independence truth conditions.
2. Different sets of sense relations
3. Distinct "conceptual cores"
4. Independent metaphorical extensions or proper names (Donald Duck, Peking Duck) |
|
|
Term
| These are seven pieces of evidence for the unity of facets (NOT polysemy). |
|
Definition
Evidence for unity of facets (NOT polysemy):
1. Prototypical co-occurence
2. Joint compositional properties
3. Lack of antagonism/serial composition without zeugma
4. Joint lexical relations
5. Joint extensions
6. Global reference
7. Joint nameability (we mean ALL facets usually, at the same time) |
|
|
Term
| These are some of what Grice (1975) contributed to semantics. |
|
Definition
|
|
Term
| These are Grice's (1975) Cooperative Principle and Maxims of conversation. |
|
Definition
Cooperative principle: Make your contribution such as is required...just...cooperate!
Maxims:
Quality: make your contribution true; don't say what you know to be false; don't say that for which you lack adequate evidence.
Quantity: be as informative as required, but not more.
Relation: be relevant.
Manner: Be perspicuous; avoid obscurity, ambiguity; be brief; be orderly.
flout=communicate by breaking a maxim
violate=basically not following the maxims, but not for the purpose of added meaning. |
|
|
Term
| This is what Grice (1975) had to say about CONVERSATIONAL IMPLICATURES. |
|
Definition
Conversational implicature: A speaker who says P conversationally implicates the proposition Q iff:
i. It is assumed speaker is observing the maxims and/or cooperative principle
ii. Speaker believes Q to be true
iii. Speaker
-believes hearer knows this
-expects hearer to know that speaker believes it. |
|
|
Term
| According to Levinson (1983), these are the qualities of CONVERSATIONAL IMPLICATURES. |
|
Definition
Defeasable
Suspendable
Nondetachable
Calculable
Nonconventional
Indeterminate
Reinforceable |
|
|
Term
| Levinson (1983) provides the following characteristics of implicatures. |
|
Definition
Defeasible
Suspendable
Nondetachable
Calculable
Nonconventional
Indeterminate
Reinforceable
(He also used the term "illocutionary modifiers" for words like 'frankly', etc. and talked about PRIMARY vs. LITERAL acts.) |
|
|
Term
| Austin (1961) contributed these notions to semantics. |
|
Definition
-Speech Act theory
-indirect speech acts
-performative analysis
-Felicity conditions
-performative adverbs ('hereby') |
|
|
Term
| Austin (1961) posited the following Felicity Conditions. |
|
Definition
Procedural (everything must be in place)
Execution (must be done fully/correctly)
Intentional (must have requisite thoughts, feelings, etc.)
To violate Procedural or Execution is a MISFIRE.
To violate Intentional is ABUSE. |
|
|
Term
| What does Austin (1961) say about truth value? |
|
Definition
| Speech acts are neither true nor false. Hence, the Felicity Conditions to determine the success of an utterance. |
|
|
Term
| Searle (1969) modified Austin's felicity conditions into these FOUR. |
|
Definition
Propositional content (state of affairs)
Preparatory condition (background circumstances/knowledge)
Sincerity condition (belief, intention, desire...)
Essential (what the speech act counts as) |
|
|
Term
| Kearns (2000) contributed these notions to semantics. |
|
Definition
Classes of quantifiers
Tense and aspect |
|
|
Term
| This is what Kearns (2000) had to say about (classes of) quantifiers. |
|
Definition
Quantifier: the semantic value of an NP which contains a determiner such as ALL, SOME, NO, EVERY, FEW, THREE... It also refers to the semantic value of words like NOBODY, NOTHING, SOMETHING, SOMEONE, EVERYONE...
These determiners = QUANTIFIER DETERMINERS
These NPs = QUANTIFIER EXPRESSIONS.
Classes of quantifiers: (Kearns 2000)
Proportional (all, every, most, few) are asymetrical.
Cardinal (no, some, many, several) tell about cardinality of an intersection |
|
|
Term
| In terms of classes of quantifiers, this is how Kearns (2000) distinguished two classes. |
|
Definition
Propositional (most, few...) = STRONG
Cardinal (some, no, several) = WEAK
Kroeger says to treat 'the' as a universal (strong) quantifier that specifies a cardinality of |1| when modified noun is sg. |
|
|
Term
| Partee (1995) offered this notion to semantics. |
|
Definition
| Intentions and possible worlds |
|
|
Term
| Faller (1995?) is known for working in Quechua on these. |
|
Definition
Evidentials
Evidentials are not part of the propositional content, but are illocutionary modifiers that ONLY AFFECT THE FELICITY CONDITIONS of an utterance.
Evidence:
1. They take scope over negatives
-they can't be negated "Fortunately, it isn't raining"
-negation doesn't affect outside of propositional content.
2. They don't contribute to truth conditions.
3. They can't be embedded.
4. They can't occur in Y/N questions. They can (but ambiguous) in content questions. "Honestly, who did Pilar visit" |
|
|
Term
| Talmy (1985) offered this to semantics |
|
Definition
Motion events and results
Manner of motion verbs in English have PP for direction ('to the island')
In Spanish: have a main verb direction ('went to island' swimming)
-generally limit themselves to one (or no) PP source, goal, etc.
-and do not seem to compensate with more verbs either.
So English=more explicit and Spanish=more implicit. |
|
|
Term
| Keenan (1985) discussed passives. |
|
Definition
All languages which have a passive construction allow for the agentless passive. (Keenan 1985)
In Latvian and Classical Arabic, passive agent is simply NOT on option. |
|
|
Term
| They came up with the Accessibility Hierarchy |
|
Definition
Keenan & Comrie (1977)
SUBJ OBJ OBJ2 OBL POSS OBJ of comparison
(gap<-- -->pronoun retention)
-(relativization strategies are not changed by the addition of a relativizer, like English with gap and 'that') |
|
|
Term
|
Definition
Ergativity (langauge: Dyirbal)
(Morphological ergativity is more common than syntactic ergativity) |
|
|
Term
| Mill (1843) contributed these notions to semantics. |
|
Definition
denotation and connotation
proper names (that they only denote)
his view of propositions (that they are what a sentence expresses, they are the objects of belief, and they are what can be T or F) |
|
|
Term
| Inconsistent/incompatible propositions come in two flavors. |
|
Definition
Contraries: may not be simultaneously true, but MAY be simultaneously false.
Contradictories: have opposite truth values; polar; one must be true and the other false. |
|
|
Term
Note: Some relations to remember:
entailment
contrary
contradictory
paraphrase
independent |
|
Definition
|
|
Term
Note: Logic and inference
In assessing inferences, we are asking what leads to true conclusions. The principle is that if you start from true premises and use valid rules of deduction, the conclusion should be true. |
|
Definition
| One advantage for using formal logic as the metalanguage is that the rules for drawing valid inferences in that system are very well understood. |
|
|
Term
**Review tautologies.
commutativity
associativity
De Morgan's Laws |
|
Definition
|
|
Term
| This is how a presupposition is similar to an entailment. |
|
Definition
| they both involve an automatic relationship seemingly free of contextual effects and requiring no reasoning. |
|
|
Term
| There are 2 basic approaches to PRESUPPOSITIONS. |
|
Definition
1. Truth relations about the world--to view sentences as external objects whose meaning is an attribute of them rather than something constructed by the participants. Semantics then would consist in relating one sentence-object to another and to the real world.
2. To view sentences as the utterance of individuals engaged in communicative acts, modeling the strategies speakers and hearers use to understand each other... (presupposition = inference hearer makes based on what speaker says) |
|
|
Term
| This is how a presupposition is different from an entailment. |
|
Definition
Unlike entailment, presupposition is preserved even when the presupposing clause is negated.
John's brother just got back from Texas. (John has a brother)
John's brother hasn't come back from Texas. (John still has a brother.
This doesn't work for entailment. if you negate the entailing sentence, the entailment fails:
I saw my father today.
Entail: I saw someone today.
Negated: I didn't see my father today.
Entail: maybe or may not have seen anyone. |
|
|
Term
Review truth table for presupposition.
(not that if the presupposition is false, then the proposition is neither true nor false (Frege), or is just false (Russell) |
|
Definition
|
|
Term
| This is how PRESUPPOSITION FAILURE relates to truth value. |
|
Definition
Presupposition failure presents problems for the truth-relation view, but not for the second (interactional) model.
The question is what happens to the truth value of the presupposing sentence when it involves a proper name for which there is no referent.
a. Ronald is a vegetarian.
b. Ronald exists.
a. The king of France is bald.
b. There is a king of France.
If the (b)s are untrue, the truth value of the sentences (a) are dubious--not true OR false. It's a gray area. This problem for the truth-based approach is known as the TRUTH-VALUE GAP. |
|
|
Term
| This was Bertrand Russell's solution to the truth-value gap. |
|
Definition
Russell's solution was to analyze definite descriptions (king of France) as complex expressions:
-The king of France is bald iff:
-at least 1 thing is the king
-at most 1 thing is the king
-whatever is the king is bald
This has the advantage of getting rid of the gray area, but it creates a large discrepancy between the surface langauge and the semantic representation. Does a simple statement really cover three underlying statements?
In the INTERACTIONAL MODEL, such definites are governed by conventions about the accessibility of the referents to the listener. a definite introduced without adequate preparation (presupp. failure) prompts a general question like "who/what is...?" |
|
|
Term
| Presuppositions are usually triggered by certain words or constructions. |
|
Definition
Definite nominals (the king of France)
Cleft (wh-) constructions (what she did was...)
Pseudo (it-)clefts (it was Ruby who...)
Adverbial clauses (before you learned to walk...)
Comparative clauses (even more gullible than...) |
|
|
Term
| In terms of presupposition triggers, lexical triggers can be these. |
|
Definition
1. Factive verbs (regret, realize...)
2. Some verbs of judgement (blame...)
3. Change of state verbs (start, begin, stop...) |
|
|
Term
| Another problem for the truth-based view of presuppositions is DEFEASIBILITY. |
|
Definition
When our knowledge about the world cancels presuppositions that might otherwise be made:
a. She [died/cried] before she finished her thesis.
b. Depending on the matrix verb, the presupp. will or won't hold.
If we used 'crid' instead of 'died' then the presupp. stands. So presuppositions are seen to depend on the context of our knowledge to some extent. |
|
|
Term
| Pseudo-clefting creates different presuppositions, depending on the order of constituents. |
|
Definition
It was Harry who Alice loved (she loved someone)
It was Alice who loved Harry (someone loved him)
Also, INTONATION patterns can create different presuppositions:
Alice loved HARRY.
ALICE loved Harry. |
|
|
Term
| In terms of presuppositions, this the the PROJECTION PROBLEM. |
|
Definition
A presupposition created by a simple clause does not survive when the clause is incorporated into a complex sentence.
John will regret doing linguistics (he is/will do linguistics)
If John does linguistics, he'll regret it (NO PRESUPPOSITION) |
|
|
Term
| So, three general types of contexts (pragmatic) influence presuppositions. |
|
Definition
1. Background knowledge
2. Topic of conversation (vs. focus)
3. Surrounding syntactic structures.
These problems call for PRAGMATIC theories to explain presupposition. One simple solution is to divide presuppositions into:
1. Semantic presuppositions, and
2. Pragmatic presuppositions |
|
|
Term
| Note: Other theorists maintain that presupposition is an entirely pragmatic phenomenon--part of the COMMON GROUND: the set of assumptions made by participants in a conversation, which shifts as new sentences are uttered. |
|
Definition
Accommodation: a way to cope with the fact that some presuppositions are not known to the listener.
OR
If, at time T, something is said that requires presupposition P to be acceptable, and if P is not presupposed just before T, then--ceteris paribus--presupposition P comes into existence. |
|
|
Term
| Note: Some writers integrate discussions of presupposition with traditional discourse notions like GIVEN vs. NEW information and FOCUS. Thus the same principles of relevance would govern both presuppositions and the choice of different word orders and/or intonations. |
|
Definition
|
|
Term
| This is what Frege said about PRESUPPOSITIONS. |
|
Definition
"One sentence presupposes another just in case the later must be true in order that the former have a truth value at all." (Strawson, Frege)
Saeed: classic tests for presupposition, as distinct from entailments: Presuppositions are preserved under negation and questioning. |
|
|
Term
| These are some things that can trigger presuppositions. |
|
Definition
a. definite descriptions
b. factive predicates
c. implicative predicates (manage to>try; forget to>intend to)
d. Aspectual predicates (Saeed's "change of state" verbs)
e. Temporal clauses/markers counterfacturals, comparative constructions, and some others.
f. Wh- (question) words (but not in rhetorical questions)
Presuppositions create traps: the only way to reject them is to reject the entire proposition due to offense taken ("have you stopped beating your wife yet?") |
|
|
Term
| This is how the SIL glossary defines presupposition. |
|
Definition
A background belief relating to an utterance, that
-must be mutually known or assumed by Speaker and Hearer (part of the common ground) for utterance to be appropriate
-generally will remain a necessary assumption whether the utterance is assertion/denial/question...
-can generally be associated with a specific lexical item or grammatical feature (a trigger) |
|
|
Term
| Note: Bertrand Russell held that presuppositions were part of what the proposition was asserting, BUT NOW, most people disagree. Even when asking questions (i.e., when nothing is being asserted) there can still be presuppositions. |
|
Definition
| So there are two main current approaches to presuppositions: semantic and pragmatic. |
|
|
Term
| This is an overview of the SEMANTIC view of presuppostions. |
|
Definition
Frege (1882), Strawson (1950, 1952) - a presuppostion must be true in order for the utterance to be either true or false.
Presupposition failure - if a presupposition is untrue, utterance has no truth value (truth value gap)
(more recent analyses might say that instead of presupp. failure, it is 'uninterpretable')
The truth value gap of the semantic analysis of presuppositions has lead to a move toward the PRAGMATIC analysis. |
|
|
Term
| This is an overview of the PRAGMATIC view of presuppositions. |
|
Definition
Austin (1962), Stalkner (1973, 1974), Karttunen (1974) - propositions the speaker presents as being non-controversial or whose truth the speaker takes for granted (or SEEMS to take for granted)
If it isn't part of the common grount, it may be accommodated (accepted as if it were true)
Assumed by Speaker to be non-controversial
Presupposition failure = NOT part of the common ground AND cannot be accommodated (pragmatically inappropriate or "infelicitous") |
|
|
Term
Note: Here's more about the pragmatic view of presuppositions:
What drives presuppositions, then, are appropriateness or Felicity Conditions.
Cleft sentences are well known for including presuppositions. These are FOCUS-PRESUPPOSITION constructions. |
|
Definition
There is special use of the cleft pattern, often in writing, where new information is included in the presuppostion (subordinate clause). These sentences have been called "informative presupposition clefts" (Prince (1978)--made possible by a kind of pragmatic accommodation by the hearer.
Ex: "It was just about 50 years ago that Henry Ford gave us the weekend." |
|
|
Term
Distinguishing PRESUPPOSITION from ENTAILMENT:
1. Presuppositions are sensitive to context and can be cancelled by certain contexts.
2. Entailments cannot be cancelled/negated; presuppositions can be negated ONLY WHEN also negating the clause containing the trigger.
3. Presupposing |
|
Definition
Examples:
1. (cried vs. died)
2. (often requires special (metalinguistic) negation (in English, intonation)).
Entail:#John killed the wasp, but the wasp isn't dead.
Presup:John didn't MANAGE to solve the problem, he solved it easily.
3. a. If John majors in linguistics, he will regret it. (doesn't presuppose he did)
b. Either John will major in linguistics, or he will regret that he did not. (does not presuppose that he did OR didn't) |
|
|
Term
| Conventional implicatures |
|
Definition
-part of the word meaning itself, but does not affect truth conditions.
To defeat implicatures, you add an additional clause to negate the implicature:
a. I'm out of petrol.
b. There's a garage round the corner...Oh, but they're out of petrol too. (implicature = cancelled)
Suspended implicature: There's a garage around the corner...but I'm not sure if they have petrol.
**Entailments and conversational implicatures go away when the initial statment is negated. Presuppositions do not.
**Top scholars in the field cannot distinguish conventional implictures from presuppositions. |
|
|
Term
| A bit about Speech Acts... |
|
Definition
Speech acts are what the speaker is trying to accomplish, such as ask a question (getting information), offer an invitation, etc.
2 Important Characteristics:
1. Interactivity- The speaker is in coordinated activity with other speakers--a no-response prompts compensation, like repeating the question.
2. Context Dependence- Institutional facts: a special context where words have special meaning only when spoken by the right people in the right circumstance, with the right authority. (judge, pastor, police officer...)
Local context: a speech act may signal different things in different contexts. (using a question to make a statement) |
|
|
Term
| We can identify sentence types when grammatical form and speech act function match. |
|
Definition
1. Declarative...assertions
2. Interrogative...questions
3. Imperative...commands
4. Optative...wishes |
|
|
