Updated 09 November
2000
The two tests for validity
entailment - what the premises do to the conclusion in a valid argument
1.
semantic entailment
semantic turnstile
test via truth-tables
P1, P2, P3 |=
C
2.
syntactic entailment
syntactic turnstile
test via tableau
P1, P2, P3 |- C
3.
The Big Surprise
P1, P2, P3 |= C
just in case
P1, P2, P3 |- C
4.
Semantics = meaning
Syntax
P Q | P and Q
--------------------------------------------------
T T | T
T F | F
F T | F
F F | F
P Q | P or Q
---------------------------------------------------
T T | T
T F | T
F T | T
F F | F
5.
G: God is all good.
P : God is all powerful.
E : There is evil.
[[G Ù P] ® ØE], E |= [ØG Ú ØP]
G P E | [[GÙP]®ØE] | E || [ØGÚØP]
-------------------------------------------------------------------
T T T | T F F |
T || F F F
T T F | T T T | F ||
F F F
T F T | F T F | T ||
F T T
T F F | F T T | F ||
F T T
F T T | F T F |
T || T T F
F T F | F T T | F ||
T T F
F F T | F T F | T ||
T T T
F F F | F T T |
F || T T T
6.
CES
{ [[G Ù
P] ®ØE],
E, Ø [ØG
Ú ØP]
}
[[G
Ù P] ® ØE]
E
Ø[ ØG Ú ØP]
|
ØØG
ØØP
Ø[G Ù P] ØE
ØG ØP
tableau closed
ces inconsistent
rgument valid
7.
Why bother to do it twice?
truth-tables more basic
why not stick to them ?
not applicable to more sophisticated logical language
need a proof technique
alternatives -
natural deduction
axiomatic method
8.
P : All dons are boring.
Q : Icabod is a don.
R : Icabod is boring.
P, Q |- R
CES { P, Q, ØR}
tableau
P
Q
ØR
9.
Inside the Proposition
Icabod is happy.
Clinton is funny.
William is a loser.
10.
1. Referring
picking
out something
Icabod,
for instance
2. Predicating
using a
predicate to
ascribe a
property
“ is
happy”
11.
Referring Devices
1. Proper Names
“William
Hague”
2. Pronouns
“She”
3. Demonstratives
“This”,
“That”
4. Definite Descriptions
“The rich
student”
“The best
logician”
Hodges : designators
12.
Designators
Hodges includes non-count nouns
butter
beer
bacon
Many are words for stuff
13.
Primary Reference
“When a designator can be used on its own in a situation so as
to refer to something, we call that thing the primary reference of the
designator”
Hodges p. 156
14.
Primary Reference
shopping list approach
primary reference is what
you would fetch if I gave you a bare list of designators
15.
Margaret Thatcher
The President of OUSU
The fastest car in Oxford
Superman
NS’s list of designators
The Master of Balliol
10391
16.
“The Master of Balliol”
fetch me that new
Labour economist who presides over the only college more ugly than Keble
also known as Andrew
Graham
“The Master of Balliol
was in love with Florence Nightingale”
17.
Andrew Graham was never in love with Florence Nightingale.
Benjamin Jowett was however.
boring correspondence
“The Master of Balliol” does not always designate its primary
reference.
18.
Purely Referential
occurrence of a Designator
Designator refers to its primary referent
19.
Purely Referential
“In fact there are many roles a designator may play in a sentence,
besides referring to its primary reference. Sometimes it refers to something
else; sometimes it doesn’t refer to anything at all.”
Hodges p.156
20.
Policy 1 on Reference
Restrict attention to designators that have a primary reference.
No vacuous names.
Ignore names that do not stand for anything
Gandalf
Cerberus
Superman
21.
Policy 2 on reference
Designators designate their primary references.
If they do not, we ignore them.
Hodges puts this more obscurely.
22.
Another restriction
we will focus on things and ignore stuff
Things can be counted -
students, trees, Porsches
You cannot count stuff
coffee, snow, butter
Stuff is picked out by non-count nouns.
23.
Proper Names
m, n, o, ...
n : Icabod
n is happy.
24.
Predicates
Icabod is happy.
----- is happy
Isabel is smarter than
Icabod.
---is smarter than ~~~
use variables to show the blanks
x is happy
x is smarter than y
25.
Hn : Icabod is happy.
m: Isabel
Smn
Predicates that need two or more names are called relational predicates.
26.
Hodges
“ a string of words and
variables such that if you replace the variables with names the whole is a
declarative sentence”
27.
More complex predicates
Reading is between Oxford and London.
n: Reading
o : Oxford
m : London
Bxyz : x is between y
and z
Bnom
28.
Quantifiers
another way of forming an indicative sentence from a predicate
“Everyone”
“Everyone is happy”
29.
Quantifiers have a domain which must be specified.
Quantifiers indicate something about number
of things in the domain
30.
All
Some (someone)
No one
At least one
Exactly two
At most three
~~~~~~~~~~~~~~~~
Most
Few
Many
31.
Universal Quantifier
All
"x___
"x
+ Hx
"xHx
For all x, Hx
Take anything you like (in the domain) it is such that it is H.
32.
Nobody is happy.
n: nobody
Hx : x is happy
Hn
We are all sad but at least Nobody is happy.
"xØHx
Take anyone you like in the domain, she or he is not happy.
33.
Existential Quantifier
$x
Hx
$xHx
You can find something (someone) such that it is happy.
34.
Domain : persons in this room
Lxy : x loves y
n : Icabod
m: Isabel
Lnn
Lmn
Lnm
Lmm
35.
Lxy : x loves y
n : Icabod
$xLnx
$xLxn
$x$yLxy
"xLnx
"xLxn
"x"yLxy
"y"xLxy
36.
Someone loves everyone.
$x"yLxy
Everyone is loved by someone.
"y$xLxy
These are very different !
37.
Someone loves
everyone.
$x"yLxy
b
a
c
Everyone
is loved by someone.
"y$xLxy
b
![]()
![]()
a c
These are very different !
38.
Domain : 1, 2, 3, ...
Bxy : x is bigger than y
"x$yByx
true - the numbers go on forever
$y"xByx
false - you can’t find a number which is bigger than any number!
39.
Domain : people in Oxford
Sx : x is a student
Rx : x is rich
All students are rich.
"x[Sx ® Rx]
?? "x[Sx
Ù Rx] WRONG
No students are rich
"x[Sx ® ØRx]
40.
Some students are rich.
$x[Sx Ù Rx]
Some students are not rich
$x[Sx Ù ØRx]
41.
“All” goes with ®
“Some” goes with Ù
!!! $x[Sx ® Rx]
P ® Q
ØP Ú Q
$x[ ØSx Ú Rx]
one rich don is enough even though
all students are poor!!
42.
All students are rich
"x[Sx
® Rx]
No students are rich
"x[Sx ® ØRx]
Some students are rich
$x[Sx Ù Rx]
Some students are not rich
$x[Sx Ù ØRx]
43.
"x[Sx ® Rx]
All students are rich.
Any student is rich.
Every student is rich.
Students are rich.
Each student is rich.
44.
Theorems
|- R Ú ØR
CES: { Ø[R
Ú ØR]
}
Tableau
Ø[R Ú ØR ]
ØR
ØØR
Theorems : proved without assumptions.
45.
Tautologies
R | [R Ú ØR]
------------------------------------
T | T
F
F | F T
Tautologies - true come what may.
You don’t need a weather person to ...
|=
R Ú Ø
R
46.
Something is a tautology just in case it is a
theorem
|= A if and only
if |-
A
47.
Keynes

Domain
residents of Oxford
49.