0% found this document useful (0 votes)
203 views

Solutions: Louis Barson Kyoto University February 2, 2009

This document contains Louis Barson's solutions to various logic problems involving modal operators such as necessity () and possibility (♦). It includes proofs showing the validity or invalidity of inferences through the use of tableau methods and countermodels. For inferences found to be invalid, a countermodel is provided consisting of a possible worlds diagram.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
203 views

Solutions: Louis Barson Kyoto University February 2, 2009

This document contains Louis Barson's solutions to various logic problems involving modal operators such as necessity () and possibility (♦). It includes proofs showing the validity or invalidity of inferences through the use of tableau methods and countermodels. For inferences found to be invalid, a countermodel is provided consisting of a possible worlds diagram.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Solutions

Louis Barson
Kyoto University
February 2, 2009

1. Prove that the truth value of ¬A at a world is the same as that of ♦¬A.

vw (¬A) = 0

iff vw (A) = 1
iff for all w′ such that wRw′ , vw′ (A) = 1
iff for all w′ such that wRw′ , vw′ (¬A) = 0
iff vw (♦¬A) = 0

2. Show the following. Where the tableau does not close, use it to define a
countermodel, and draw this, as in 2.4.8

(a) ⊢ (A ∧ B) ⊃ (A ∧ B)

(A ∧ B), 0
¬(A ∧ B), 0
A, 0
B, 0
♦¬(A ∧ B), 0
0r1
¬(A ∧ B), 1

¬A, 1 ¬B, 1
A, 1 B, 1
⊗ ⊗

1
(b) ⊢ (A ∧ B) ⊃ (A ∨ B)

(A ∨ B), 0
¬(A ∨ B), 0
♦¬(A ∨ B), 0
0r1
¬(A ∨ B), 1
¬A, 1
¬B, 1

A B
A, 1 B, 1
⊗ ⊗

(c) ⊢ A ≡ ¬♦¬A

¬(A ≡ ¬♦¬A), 0

A, 0 ¬A, 0
¬¬♦¬A, 0 ¬♦¬A, 0
♦¬A, 0 ♦¬A, 0
¬A, 0 ⊗

(d) ⊢ ♦A ≡ ¬¬A

¬(♦A ≡ ¬¬A), 0

♦A, 0 ¬♦A, 0
¬¬¬A, 0 ¬¬A, 0
¬A, 0 ¬A, 0
¬♦A, 0 ⊗

2
(e) ⊢ ♦(A ∧ B) ⊃ (♦A ∧ ♦B)

¬(♦(A ∧ B) ⊃ (♦A ∧ ♦B)), 0


♦(A ∧ B), 0
¬(♦A ∧ ♦B), 0

¬♦A, 0 ¬♦B, 0
¬A, 0 ¬B, 0
0r1 0r1
A ∧ B, 1 A ∧ B, 1
A, 1 A, 1
B, 1 B, 1
¬A, 1 ¬B, 1
⊗ ⊗

(f) ⊢ ♦(A ∨ B) ⊃ (♦A ∨ ♦B)

¬(♦(A ∨ B) ⊃ (♦A ∨ ♦B)), 0


♦(A ∨ B), 0
¬(♦A ∨ ♦B), 0
¬♦A, 0
¬♦B, 0
¬A, 0
¬B, 0
0r1
A ∨ B, 1

A, 1 B, 1
¬A, 1 ¬B, 1
⊗ ⊗

3
(g) (A ⊃ B) ⊢ ♦A ⊃ ♦B

(A ⊃ B), 0
¬(♦A ⊃ ♦B), 0
♦A, 0
¬♦B, 0
¬B, 0
0r1
A, 1
A ⊃ B, 1
¬B, 1

¬A, 1 B, 1
⊗ ⊗

(h) A, ♦B ⊢ ♦(A ∧ B)

A, 0
♦B, 0
¬♦(A ∧ B), 0
¬(A ∧ B), 0
0r1
B, 1
A, 1
¬(A ∧ B), 1

¬A, 1 ¬B, 1
⊗ ⊗

(i) ⊢ A ≡ (¬A ⊃ A)

¬(A ≡ (¬A ⊃ A)), 0

A, 0 ¬A, 0
¬(¬A ⊃ A), 0 (¬A ⊃ A), 0
♦¬(¬A ⊃ A), 0 ♦¬A, 0
0r1 0r1
¬(¬A ⊃ A), 1 ¬A, 1
A, 1 ¬A ⊃ A, 1
¬A, 1
¬A, 1 ¬¬A, 1 A, 1
⊗ A, 1 ⊗

4
(j) ⊢ A ⊃ (B ⊃ A)

¬(A ⊃ (B ⊃ A)), 0


A, 0
¬(B ⊃ A), 0
♦¬(B ⊃ A), 0
0r1
¬(B ⊃ A), 1
B, 1
¬A, 1
A, 1

(k) ⊢ ¬♦B ⊃ (B ⊃ A)

¬(¬♦B ⊃ (B ⊃ A)), 0


¬♦B, 0
¬(B ⊃ A), 0
♦¬(B ⊃ A), 0
¬B, 0
0r1
¬(B ⊃ A), 1
B, 1
¬A, 1
¬B, 1

5
(l) 0 (p ∨ q) ⊃ (p ∨ q)

¬((p ∨ q) ⊃ (p ∨ q)), 0


(p ∨ q), 0
¬(p ∨ q), 0
¬p, 0
¬q, 0
♦¬p, 0
♦¬q, 0
0r1
¬p, 1
p ∨ q, 1

p, 1 q, 1
⊗ 0r2
¬q, 2
p ∨ q, 2
p, 2 q, 2

The following interpretation shows this inference to be invalid:

W = {w0 , w1 , w2 }

w0 Rw1 , w0 Rw2
vw1 (p) = 0, vw1 (q) = 1, vw2 (p) = 1, vw2 (q) = 0

This can be represented in the following diagram:

w0

w1 w2
¬p, q p, ¬q

6
(m) p, ¬q 0 (p ⊃ q)

p, 0
¬q, 0
¬(p ⊃ q), 0
♦¬(p ⊃ q), 0
0r1
¬(p ⊃ q), 1
p, 1
¬q, 1
p, 1
¬q, 1

The following interpretation shows this inference to be invalid:

W = {w0 , w1 }

w0 Rw1
vw1 (p) = 1, vw1 (q) = 0

This can be represented in the following diagram:

w0

w1
p, ¬q

7
(n) ♦p, ♦q 0 ♦(p ∧ q)

♦p, 0
♦q, 0
¬♦(p ∧ q), 0
¬(p ∧ q), 0
0r1
p, 1
¬(p ∧ q), 1

¬p, 1 ¬q, 1
⊗ 0r2
q, 2
¬(p ∧ q), 2

¬p, 2 ¬q, 2

The following interpretation shows this inference to be invalid:

W = {w0 , w1 , w2 }

w0 Rw1 , w0 Rw2
vw1 (p) = 1, vw1 (q) = 0, vw2 (p) = 0, vw2 (q) = 1

This can be represented in the following diagram:

w0

w1 w2
p, ¬q ¬p, q

8
(o) 0 p ⊃ p

¬(p ⊃ p), 0
p, 0
¬p, 0

The following interpretation shows this inference to be invalid:

W = {w0 }

vw0 (p) = 0

This can be represented in the following diagram:

w0
¬p

(p) 0 p ⊃ ♦p

¬(p ⊃ ♦p), 0
p, 0
¬♦p, 0
¬p, 0

The following interpretation shows this inference to be invalid:

W = {w0 }

This can be represented in the following diagram:

w0

9
(q) p 0 p

p, 0
¬p, 0
♦¬p, 0
0r1
¬p, 1

The following interpretation shows this inference to be invalid:

W = {w0 , w1 }

w0 Rw1
vw0 (p) = 1, vw1 (p) = 0

This can be represented in the following diagram:

w0 p

w1
¬p

(r) 0 p ⊃ p

¬(p ⊃ p), 0
p, 0
¬p, 0
♦¬p, 0
0r1
¬p, 1
♦¬p, 1
p, 1
1r2
¬p, 2

The following interpretation shows this inference to be invalid:

W = {w0 , w1 , w2 }

w0 Rw1 , w1 Rw2
vw1 (p) = 1, vw2 (p) = 0

10
This can be represented in the following diagram:

w0

w1 p

w2
¬p

(s) 0 ♦p ⊃ ♦♦p

¬(♦p ⊃ ♦♦p), 0
♦p, 0
¬♦♦p, 0
¬♦p, 0
0r1
p, 1
¬♦p, 1
¬p, 1

The following interpretation shows this inference to be invalid:

W = {w0 , w1 }

w0 Rw1
vw1 (p) = 1

This can be represented in the following diagram:

w0

w1
p

11
(t) 0 p ⊃ ♦p

¬(p ⊃ ♦p), 0
p, 0
¬♦p, 0
♦¬♦p, 0
0r1
¬♦p, 1
¬p, 1

The following interpretation shows this inference to be invalid:

W = {w0 , w1 }

w0 Rw1
vw0 (p) = 1

This can be represented in the following diagram:

w0 p

w1

(u) 0 ♦p ⊃ ♦p

¬(♦p ⊃ ♦p), 0
♦p, 0
¬♦p, 0
♦¬♦p, 0
0r1
p, 1
0r2
¬♦p, 2
¬p, 2

The following interpretation shows this inference to be invalid:

W = {w0 , w1 , w2 }

w0 Rw1 , w0 Rw2
vw1 (p) = 1

12
This can be represented in the following diagram:

w0

w1 w2
p

(v) 0 ♦(p ∨ ¬p)

¬♦(p ∨ ¬p), 0
¬(p ∨ ¬p), 0

The following interpretation shows this inference to be invalid:

W = {w0 }

This can be represented in the following diagram:

w0

4. *Check the details omitted in 2.9.3 and 2.9.6.

2.9.3 Soundness Lemma: Let b be any branch of a tableau, and I = < W, R, v >
be any interpretation. If I is faithful to b, and a tableau rule is applied to it,
then it produces at least one extension, b′ , such that I is faithful to b′ .

Proof:

The proof proceeds by a case-by-case consideration of the tableau rules. A ∨ B ,


A ⊃ B , A ≡ B , ¬¬A , and ¬A have not been explicitly dealt with.

A∨B
Suppose I is faithful to b and A ∨ B, i occurs on b, and that we apply a rule to
it. Then two branches eventuate - one extending b with A, i (the left branch)
and one extending b with B, i (the right branch). Since I is faithful to b it
makes every formula on b true - in particular A ∨ B is true at f (i), that is,
vwi (A ∨ B) = 1, so either A or B is true at f (i): i.e. vwi (A) = 1 or, vwi (B) = 1.

13
In the first case, I is faithful to the left branch; in the second case I is faithful
to the right branch.

Suppose I is faithful to b and ¬(A ∨ B), i occurs on b, and that we apply a rule
to it. Then one branch eventuates - extending b with ¬A, i and¬B, i. Since I is
faithful to b it makes every formula on b true - in particular vwi (¬(A ∨ B)) = 1
so vwi (A) = 0 and, vwi (B) = 0, making I faithful to the extended branch.

A≡B
Suppose I is faithful to b and A ≡ B, i occurs on b, and that we apply a rule to
it. Then two branches eventuate - one extending b with A, i and B, i (the left
branch) and one extending b with ¬A, i and¬B, i (the right branch). Since I is
faithful to b it makes every formula on b true - in particular vwi (A ≡ B) = 1 so
vwi (A) = vwi (B). That is, either vwi (A) = 1 and vwi (B) = 1 or, vwi (A) = 0
and vwi (B) = 0. In the first case, I is faithful to the left branch, in the second
case I is faithful to the right branch.

Suppose I is faithful to b and ¬(A ≡ B), i occurs on b, and that we apply a rule
to it. Then two branches eventuate - one extending b with A, i and ¬B, i (the
left branch) and one extending b with ¬A, i B, i (the right branch). Since I is
faithful to b it makes every formula on b true - in particular vwi (¬(A ≡ B)) = 1
so vwi (A) 6= vwi (B). That is, either vwi (A) = 1 and vwi (B) = 0 or, vwi (A) = 0
and vwi (B) = 1. In the first case, I is faithful to the left branch, in the second
case I is faithful to the right branch.

¬¬A
Suppose I is faithful to b and ¬¬A, i occurs on b, and that we apply a rule to it.
Then just one branch eventuates: extending b with A, i. Since I is faithful to b
it makes every formula on b true - in particular vwi (¬¬A) = 1 so vwi (A) = 1.
Thus I is faithful to the extended branch.

¬A
Suppose I is faithful to b and ¬A, i occurs on b, and that we apply a rule to it.
Then just one branch eventuates: extending b with ♦¬A, i. Since I is faithful
to b it makes every formula on b true - in particular vwi (¬A) = 1. By the
proof in question 1, this implies that vwi (♦¬A) = 1. Thus I is faithful to the
extended branch.

14
2.9.6 Finish the proof of the Completeness Lemma: The atomic case, B ∨ C,
B, and ¬B have already been shown.

Completeness Lemma: Let b be any open complete branch of a tableau. Let


I =< W, R, v > be the interpretation induced by b. Then:

if A, i is on b then A is true at wi
if ¬A, i is on b then A is false at wi

Proof:
The proof is by recursion on the complexity of A.

¬(B ∨ C)

If A occurs on b, and is of the form ¬(B ∨ C), then the rule for disjunction
has been applied to ¬(BvC), i. Thus, ¬B, i and ¬C, i are on b. By induction
hypothesis, both ¬B and ¬C are true at wi . Hence (B ∨ C) is false at wi , as
required.

B ∧C

If A occurs on b, and is of the form B ∧ C, then the rule for conjunction has
been applied to B ∧ C, i. Thus, B, i and C, i are on b. By induction hypothesis,
both B and C are true at wi . Hence B ∧ C is true at wi , as required.

¬(B ∧ C)

If A occurs on b, and is of the form ¬(B ∧ C), then the rule for conjunction
has been applied to ¬(B ∧ C), i. Thus, ¬B, i or ¬C, i is on b. By induction
hypothesis, either ¬B or ¬C is true at wi . Hence B ∧ C is false at wi , as
required.

B⊃C

If A occurs on b, and is of the form B ⊃ C, then the rule for the conditional has
been applied to B ⊃ C, i. Thus, ¬B, i or C, i is on b. By induction hypothesis,
either ¬B or C is true at wi . Hence B ⊃ C is true at wi , as required.

¬(B ⊃ C)

If A occurs on b, and is of the form ¬(B ⊃ C), then the rule for the conditional
has been applied to ¬(B ⊃ C), i. Thus, B, i and ¬C, i are on b. By induction
hypothesis, both B and ¬C are true at wi . Hence B ⊃ C is false at wi , as
required.

15
B≡C

If A occurs on b, and is of the form B ≡ C, then the rule for equivalence has
been applied to B ≡ C, i. Thus, B, i and C, i, or ¬B, i and ¬C, i are on b. By
induction hypothesis, either B and C, or ¬B and ¬C are true at wi . Hence
B ≡ C is true at wi , as required.

¬(B ≡ C)

If A occurs on b, and is of the form ¬(B ≡ C), then the rule for equivalence has
been applied to ¬(B ≡ C), i. Thus, B, i and ¬C, i, or ¬B, i and C, i are on b.
By induction hypothesis, either B and ¬C, or ¬B and C are true at wi . Hence
(B ≡ C) is false at wi , as required.

¬¬B

If A occurs on b, and is of the form ¬¬B, then the rule for double negation has
been applied to ¬¬B, i. Thus, B, i is on b. By induction hypothesis, B is true
at wi . Hence ¬B is false at wi , as required.

♦B

If A occurs on b, and is of the form ♦B, then the rule for possibility has been
applied to ♦B, i. Thus, for some j such that iRj is on b, B, j is on b. By
construction and the induction hypothesis, for some wj such that wi Rwj , B is
true at wj . Hence ♦B is true at wi , as required.

¬♦B

If A occurs on b, and is of the form ¬♦B, then the rule for possibility has been
applied to ¬♦B, i. Thus, ¬B, i is on b. So, for all j such that iRj, ¬B, j is on
b. By induction hypothesis, for all j such that wi Rwj , B is false at wj . Hence
♦B is false at wi , as required.


16

You might also like