CAS LX 502 Semantics

1
CAS LX 502Semantics

• 2b. Sense and concepts
• 2.4-

2
Mental representations

• We can talk about things that dont exist, things
  as they might have been, things as they will be.
  Language is not limited to references to tangible
  entities and properties in the real world.
• We have a mental representation of the meaning of
  language (and much of what we are studying here
  are the properties of these mental
  representations).

3
Images

• The mental representation of the meaning of a
  word, even a name, cannot simply be an image or
  other perceptually grounded representation
  (justice? happiness?). Rather, we have a complex
  mental concept as the meanings of wordstheir
  sense.
• Perceptual representations may be part of the
  concept, but they are not the all there is to the
  concept.

4
Concepts transcend words

• Of course, there need not be a word for there to
  be a concept. Concepts can be described, and
  given a designation if they are useful.

• sniglet n. Any word that doesnt appear in the
  dictionary but should.
• Rich Hall had a series of books in the 80s
  containing these.
• eastroturf n. The artificial grass in Easter
  baskets.
• furnidents n. The indentations that appear in
  carpets after a piece of furniture has been
  removed.

5
What is our concept of bird?

• When we think of the meaning of bird, the concept
  of bird, we think of it as applying to some
  things and not others. At a first pass, it might
  have to satisfy these necessary conditions
  (linking this concept to other concepts animal,
  wings, )
• It is an animal
• It has two wings
• It can fly
• It sings or chirps

6
Necessary and sufficient

• Not all of these linked concepts are really
  necessary for something to fall under the concept
  bird.
• Is a penguin a bird? Surely.Yet it doesnt fly.
  Is a fish with three eyes really a
  fish? So what are the sufficient
  conditions?

7
According to Webster

• Bird n. 2 any of a class (Aves) of
  warm-blooded vertebrates distinguished by having
  the body more or less completely covered with
  feathers and the forelimbs modified as wings.
• Wing n. 1 a one of the movable feathered or
  membranous paired appendages by means of which a
  bird, bat, or insect is able to fly also such
  an appendage even though rudimentary if possessed
  by an animal belonging to a group characterized
  by the power of flight

8
According to Webster

• Bird n. 2 any of a class (Aves) of
  warm-blooded vertebrates distinguished by having
  the body more or less completely covered with
  feathers and the forelimbs modified as wings.
• Wing n. 1 a one of the movable feathered or
  membranous paired appendages by means of which a
  bird, bat, or insect is able to fly also such
  an appendage even though rudimentary if possessed
  by an animal belonging to a group characterized
  by the power of flight

In other words, we ask a bird expert.
Thus penguins are birds?
9
Folk semantics

• Yet, were pretty happy to call penguins birds,
  as well as ostriches. We might call a whale or a
  dolphin a fish, or we might call a tomato a
  vegetable, even if the experts would tell us
  otherwise. We may be able to discern the
  difference between an elm and a beech only by
  seeing how an expert reacts when faced with one.
• We have some kind of concept of fish, vegetable,
  bird, elm that doesnt seem to rely on a strict
  definition (necessary and sufficient conditions).

10
Prototypes

• So how is our conceptual knowledge organized?
  Perhaps a concept has a prototypical member,
  meeting all of the conditions, where things that
  meet only some of the conditions (having only
  some of the characteristic features) are more
  peripheral. The closer the match, the more
  typical.
• Some features are clearly more fundamental than
  others. A mechanical bird would seem a much less
  typical bird than an ostrichprobably not even a
  bird at all.

11
Characteristic features, prototypical exemplars

• In a sense, these ideas make intuitive sensebut
  in another sense, they are no use to us. These
  dont seem to tell us anything about how concepts
  can and cant be organized, really. They dont
  explain why concepts are the way they are. They
  just give us a language of description once we
  already know the facts.

12
The acquisition of concepts

• Any attempt to define a concept will quite
  rapidly run into a circularity problem. Concepts
  are clearly related to one another.
• Cf. conceptual networks A duck is a bird, and
  inherits from bird the conceptual associations
  bird has, such as being an animal, inheriting the
  conceptual associations animal has, etc.).
• But eventually, there must be nothing to reduce
  to. (Or is it turtles all the way down?)

13
The acquisition of concepts

• This is a problem for the philosophers, perhaps,
  but it does not seem beyond the realm of
  possibility that there are a certain number of
  fundamental concepts that we start out with
  (perhaps like the structure of language
  attributable to UG), and from which other
  concepts are derived. The philosophical
  implications are many. But we do seem, as kids,
  to know how to characterize doggie when a doggie
  is pointed out to us (cf. Quines gavagai).

14
Language, thought, and reality

• Dont you see that the whole aim of Newspeak is
  to narrow the range of thought? In the end we
  shall make thoughtcrime literally impossible,
  because there will be no words in which to
  express it. Every concept that can ever be
  needed, will be expressed by exactly one word,
  with its meaning rigidly defined and all its
  subsidiary meanings rubbed out and forgotten.
  (according to Syme, from George Orwells 1984, of
  course)

15
Sapir-Whorf

• B.L. Whorf, studying Uto-Aztecan languages,
  observed that agreement systems, declensions,
  tense-marking vary widely across languages, and
  languages often divide word into classes based on
  rather arbitrary criteria.
• He has since been much maligned for suggesting
  that these classifications impose a restriction
  on how speakers classify conceptstaken to an
  extreme, this leads to the idea of Newspeak.
  Known as the Sapir-Whorf hypothesis.

16
Pinker v. Whorf

• Pinker (The Language Instinct) runs through a
  number of arguments against the idea that
  language determines concepts.
• 5-mo olds react to unexpected changes in
  cardinality.
• Vervet monkeys sister bit VMs antagonists
  sister.
• Mental rotation 56 RPM.
• So what is the language of concepts? Not English.
  We might call it Mentalese, if we so wish, but
  whatever we call it, it is richer, more
  unambiguous, and certainly distinct from spoken
  languages.

17
? snow

• ? powder ? flurry
• ? sleet
• ? hail ? dusting
• ? blizzard ? hardpack
• ? slush
• ? avalanche