Who is interested? Linguists, psychologists, lawyers, advertisers...
Conventional sentence meaning arises from the combination of individual morpheme meanings (sense) and syntactic structure.
The meaning of expressions may have two components, sense (dictionary meaning) and referent (what the term or phrase refers to in context.) Two expressions, e.g. George Washington, the first president of the USA, may have the same referent yet different sense. (Recall Frege from Semenza & Zettin (1989)
Jung,
C. G. (1910). "The association method." American Journal of Psychology
21: 219-269.
"...it is possible to strike a concealed (indeed unconscious) complex by means
of a stimulus word...One must get rid of the idea that educated and intelligent
test persons are able to see and admit their own complexes. Every human mind
contains much that is unacknowledged and hence unconscious as such; and no one
can boast that he stands completely above his complexes. 235" Jung's method was
based on Galton (1883), who noted the ability of the method to "lay bare the
foundation..of thoughts."
Galton, F. (1883). An inquiry into human faculty and its development.
London, Dent.
hot
__ __ __ __ __ __ __ cold
Osgood et al (1957) used these 7 point scales in studies of affective meaning.
For example, he showed that the multiple personality "Eve" had different
meanings for personal concepts in her different personalities.
These studies revealed practical utility in assessing similarities and
differences between objects rated, as well as the universality of the
qualifier-quantifier system in human language. (One can construct identical
rating scales in any language. These checkpoints can be interpreted as
individual statements that the rater is selecting from., e.g.
The ______ is very hot.
The_______ is quite hot
The_______ is somewhat hot
The_______ is neither hot nor cold.
The_______ is somewhat cold.
The_______ is quite cold.
The_______ is very cold.
Remember, word/morpheme meaning is what a morpheme contributes to its sentences and we know its sentences' meanings both directly and more analytically by looking at patterns of inferences drawn by listeners from those sentences in contexts and how that pattern changes as words change.
While nearly synonyms in sense, the latter displays a certain attitude of the speaker toward the references of the phrase containing it, e.g."Nixon's henchmen/colleagues." Sometimes known as affective or connotative meaning, this is an interesting realm of meaning research in regard to advertising and product names. See Gruner et al (1991) on "doublespeak" and euphemisms, "pre-owned" vs "used" car.
Using
this involves recognizing it modifies a physical dimension of objects, usually
a comparison requiring extra-linguistic "database" information, e.g. about room
sizes, basketball, midgets, ten-year olds, etc. The comparison may be implicit
or explicit. J is tall or J is tall for a midget.
"tall" is one of those scalable adjectives that engages the complex
quantification system of English involving intensifiers (very, some...) and
quantifying complements ("so ___that [clause]"), and absolutes ("ten feet
tall").
"tall" is also one of a universal class of relative modifiers that cannot be
simply taken as an additive element to the meaning of its phrases. Thus unlike
a "brown giraffe" or a "Republican governor", a "tall dwarf" is not both tall
and a dwarf!
Describing a "gift" or whatever as tall presupposes it has a certain
shape and orientation with respect to horizontal.-- something like the normally
vertical dimension greatly exceeds the other two.
Sam
[ s] that Nixon was a great president.
What do listeners infer about the speaker as they hear these sentences? What
changes as the verb changes?
The usual idea is that of Leibniz - some primitives that are used over and over, and perhaps with each word having its own unique "gestalt" as well. Some words like tall, however, seem to be like functions or computer subroutines that gather information from the linguistic environment (nearby morphemes), local context, and the "vast database" of the listener and "compute" a complex of meaning including presuppositions, potential inferences and associations.
Many
words travel in packs -that is they have similar linguistic distribution, e.g.
tall, short, big etc. These "packs" often share meaning as
well as syntactic environments. A semantic field is a set of words that covers
the spectrum of possibilities in some domain -typically all its members have
nearly identical linguistic distribution, e.g. color names, vegetables,
relative adjectives, cognitive verbs. It is the common linguistic environment
that makes "syntactic bootstrapping" a possible strategy for determining
meaning of a novel word.
It is reasonable to suppose that the shared meaning be represented similarly.
Many schemes have been proposed, all coding word forms for similarity in meaning.
This suggests that they all have distinct if slightly different meanings even if identical linguistic distribution. Look for differences in patterns of inference as you change from one word to another.
"If every language, has so to speak, some 'one-element' words from the 'alphabet of human thoughts', expressing those basic concepts, every language has also a vast repertoire of complex concepts ('many-element words'), which constitute culture-specific configurations of the elementary building blocks and provide clues to culture specific ways of thinking." (Weirzbicka 1992, p.210)
New forms occur daily, old ones are lost, and existing ones change on many dimensions - associations, affect, even specific attributes.
This information generally comes in context but requires lexical knowledge to use that context. Thus listeners construct an interpretation of an utterance from the conventional meaning of the sentence, local context, and their "vast database" of general knowledge. Use of this general knowledge should be part of a word's meaning, even though not its sense. See the Pinker, readings on "sportugese," and euphemisms, as well as Tannen's work. (and also my notes on comprehension.)
Interpreting "Otto is tall" requires the listener supplying a reference group (for a dwarf, first-grader, NBA forward,, the average adult\ male, etc.)
See the example in comprehension notes from Bransford and Johnson (1972) on a "procedure."
These common terms are clearly relational. Their meaning depends on the context of use. Other words have similar relationships with the speaker's perspective, e.g. so-called indexicals, "right, left, behind, close", etc.
Just as the different perspectives of speaker and target audience determine choice of word and phrase in the deictic expressions, other aspects of perspective may determine expressions differing in words but not intention. Compare for example, "come in" with "go out.
Tannen points out miscommunication can occur even when words and grammar are totally shared, e.g. by husband and wife.
The stage is set on one end by phonology which establishes "possible word" schemas, and on the other by "mentalese" which establishes the initial "possible thoughts." There also must be some specific word acquisition strategies. See 10.
Children appear to have a set of procedures including the "whole-object assumption", "mutual exclusivity" etc. to connect words and things.
Natural kinds may be more like proper names with their "meaning" not a part of language but the "vast database."
(Compare
the "cloze" procedure.)
I am too ________ to play golf.
Using whatever means, children tend to grab a sense for a new word on the first exposure. This is necessary if they are to grow their vocabulary at 5-10 words a day.