PHIL
309: Twentieth Century Analytic Philosophy
LARKIN:
Spring 2003
________________________________________
I.
The
Theory of Descriptions
A.
Denoting
Phrases
1.
Indefinite
Descriptions
a.
Universal
quantifier phrases
“All F’s”
b.
Existential
quantifier phrases
“Some F”
“An F”
2.
Definite
Descriptions
“The F”
B.
Main
Thesis
1.
Denoting
Phrases (DPs) do not have meaning in isolation.
2.
DPs
do not stand for genuine constituents of propositions.
3.
DPs
are not like (genuine/logically proper) names.
4.
The
meaning of a DP is not its denotation.
5.
DP
sentences are not about the objects, if any, denoted by the DP.
6.
DP
sentences are actually about properties.
C.
Russellian
Analyses
1.
Indefinite
Descriptions
a.
universal
quantifier phrase sentences
“All F are G” => “For all x, if x is F then x is
G” =>
“If something instantiates F, then it instantiates G
as well”
b.
existential
quantifier phrase sentences
“Some F is G” => “There is an x such that x is
both F and G” =>
“There is at least one thing that co-instantiates F
and G.”
2.
Definite
Descriptions
“The F is G” => “There is an x such that x is F,
and if anything else y is F then y=x, and x is G” =>
“There is one and only one F and that thing is G”
=>
“There is something that co-instantiates
being-a-unique-F and G”
D.
The
Hard Case = Definite Descriptions
Whereas it is somewhat easy to see that universal
and existential generalizations are not really claims about objects, it is much
more counter-intuitive in the case of definite description sentences. The sentence “The president of the US is
hawkish” certainly seems to be a claim about George W. Bush. But on Russell’s analysis, it is not. It is rather a claim about the properties
being a unique president of the US and being hawkish. Thus Russell spends OD mainly arguing for his analysis of
definite description sentences.
II.
Main
Argument
P1: The
theory of descriptions (TOD) avoids the difficulties of regarding denoting
phrases as standing for genuine constituents of propositions.
1a: Avoids
difficulties with apparently empty denoting phrases.
1b: Avoids
difficulties with the relationship between meanings and denotations.
P2: TOD
can give solutions to the three semantic puzzles.
2a:
The Law of Identity
Problem
2b:
The Law of Excluded Middle
Problem
2c: The
Problem with True Negative Existentials
P3: TOD
helps resolve some important philosophical issues.
3a:
Empty and Fictional Names
3b:
Frege’s Puzzle
3c: Talking
and thinking about things with which we are not acquainted.
C: TOD
is good.
III.
Frege
and Meinong
1.
Russell
holds that DPs do not correspond to genuine constituents of propositions—DP
sentences are not about the denotation of a DP.
2.
Frege
and Meinong think that DPs do correspond to genuine constituents of
propositions—they must say that DP sentences are at least in aprt about the
denotation of the DP.
3.
The
Problem: Denoting Phrases that are apparently empty
a.
Meinong
must say that these DPs correspond to an existent object that does not subsist.
b.
Frege
must say that these denoting phrases do have a sense but no referent, but he
stills needs to have sentences involving these DPs be about something other
than the sense—so Frege must find a referent even for those sense that appear
to not have any referent.
4.
Russell
doesn’t have to worry about trying to find some referent of a denoting phrase
that appears to be empty, for he analyzes those phrases away and is not
committed to DP sentences being about the denotation of the DP.
P1: If
Meinong is right, then the phrase ‘the round square’ denotes an object that
exists but does not subsist.
P2: But if
‘the round square’ denotes an object that in some sense exists, then there is
in some sense a thing that is both round and not round at the same time.
P3: This
violates the law of non-contradiction.
C: So Meinong is wrong.
P0: Assume
that a denoting phrase expresses a meaning and denotes a denotation/referent.
P1: “The
present king of England is bald” is not about the meaning of the phrase ‘the
present king of England’ but rather its denotation.
C1: So, by parity of form, “The present King of
France is bald” is not about the meaning but the denotation of ‘the present
king of France’.
P2: If
“The present king of France is bald” is about the denotation of ‘the present
king of France’, then the sentence is nonsense (it is not about anything).
P3: But
the sentence is not nonsense (it is false [or perhaps it is neither true nor
false]).
C2: So
“The present king of France is bald” is not about the denotation of ‘the
present king of France’.
C3: Since
C1 and C2 are contradictory, the assumption (P0) must be false.
P1: If
denoting phrase sentences have a subject-predicate structure, then they are
(all) meaningful only if the denoting phrase denotes something.
P2: There
are (some) meaningful denoting phrase sentences with empty denoting phrases.
C: So denoting phrase
sentences do not have a subject-predicate structure.
Meinong objects to P2 by
claiming that apparently empty denoting phrases denote objects that exist but
do not subsist.
Russell complains that this
way around the problem violates the law of non-contradiction.
Frege objects to P3 by
claiming that apparently empty denoting phrases express senses that denote the
null class.
Russell complains that this
way around the problem is ad hoc.
IV.
Gray’s
Elegy Argument
The basic need for the
Gray’s Elegy argument arises as follows:
If DPs are to stand for genuine constituents of propositions, then (a)
they must stand for something like a meaning (Fregean sense, denoting complex)
and not just their referent [or else we would be stuck with a Meinong-style
view to account for apparently empty DPs], but (b) DP sentences still need to
be about the referent or denotation of the DP.
The Gray’s Elegy argument is intended to show that these desiderata
cannot be jointly satisfied.
Gray’s Elegy Argument (version 1)
P0: Assume:
Denoting phrases express denoting complexes that denote denotations.
P1: The
relationship between a denoting complex and the denotation must be a ‘logical
one’.
C1: So propositions expressed using denoting
phrases are solely about the denotation (they are not about the denoting
complex).
P2: “Scott is the author of Waverly”
is not solely about the denotation of the denoting phrase ‘the author of Waverly’ but also about the meaning of
that phrase (i.e., the denoting complex).
C2: So
propositions expressed using denoting phrases are not solely about the
denotation (they are also about the denoting complex).
C3: Since
C1 and C2 are contradictory the assumption must be false.
Gray’s Elegy Argument (version 2)
P1: If
denoting phrase (DP) sentences have a subject-predicate structure, then a DP
must either contribute an object (its denotation) or a denoting complex (its
meaning) to the proposition expressed by the sentence in which DP occurs.
P2: If
DP contributes an object, then the proposition expressed is solely about the
denotation and not the denoting complex.
P3: If
DP contributes a denoting complex, then still the proposition expressed is
solely about the denotation and not the denoting complex.
[Due to the logical relation that must exist between
a denoting complex and the denotation.]
P4: Propositions
expressed by DP sentences are not solely about the denotation but also in part
about the denoting complex.
[Scott/author of Waverly puzzle]
C: So DP sentences do
not have a subject-predicate structure.
V.
Semantic
Puzzles
Russell argues that his theory of descriptions can
resolve three puzzles concerning DP sentences.
He construes the puzzles as data that a semantic theory needs to account
for. They can also be seen as a test
for whether a semantic theory is consistent with the laws of logic: the law of
identity, the law of excluded middle, and the law of non-contradiction.
See ‘Semantic Puzzles’ Handout
VI.
Philosophical
Implications
A.
Empty
and Fictional Names: Difficulties here can be resolved by viewing these names
as described definite descriptions.
B.
Frege’s
Puzzle: See ‘Semantic Puzzle’ handout
C.
Knowledge
by acquaintance and knowledge by desciption:
Russell contends that we cannot speak or think about objects with which
we have had no acquaintance. Yet we
appear to do so. Russell’s theory of
description allows us to see that propositions that might appear to be about
objects with which we are not or cannot be acquainted are really propositions
about things with which we are acquainted—namely properties (and perhaps some
simple objects to which we can demonstratively refer).
VII.