For proofs on this homework,
remember that each step must be justified by one of the following:
-
a premise,
-
a
WaterWorld domain axiom,
-
a listed
inference rule
with the referenced line numbers
(and, if ambiguous, substitutions for the inference rule's meta-variables),
or
-
a subproof shown inline, or equivalently, a theorem/lemma shown
previously.
Except where otherwise directed, you may use any theorem shown in
the text or by a previous exercise, even if that exercise was not
assigned.
Fill in the blank reasons in the following proof that ∨∨ commutes,
that is,
(χ∨υ) ⊢ (υ∨χ)(χ∨υ) ⊢ (υ∨χ).
| 1 | (χ∨υ)(χ∨υ) | |
Premise
|
| 2 | subproof:χ ⊢ (υ∨χ)χ ⊢ (υ∨χ) | | |
| 2.a | | χχ |
Premise for subproof
|
| 2.b | | (υ∨χ)(υ∨χ) |
∨∨Intro, line 2.a
|
| 3 | subproof:υ ⊢ (υ∨χ)υ ⊢ (υ∨χ) | | |
| 3.a | | υυ |
Premise for subproof
|
| 3.b | | (υ∨χ)(υ∨χ) |
____________
|
| 4 | (υ∨χ)(υ∨χ) | |
____________
|
Show that
(φ∧ψ), (φ→θ), (ψ→δ) ⊢ (θ∧δ)(φ∧ψ), (φ→θ), (ψ→δ) ⊢ (θ∧δ).
It should take around 8 steps.
Show what is often called the implication chain rule:
(φ→ψ), (ψ→θ) ⊢ (φ→θ)(φ→ψ), (ψ→θ) ⊢ (φ→θ).
[Practice problem—solution provided.]
Show what is often called negated-or-elimination (left):
¬(φ∨ψ) ⊢ ¬φ¬(φ∨ψ) ⊢ ¬φ.
Think backwards. How can we end with ¬φ¬φ?
One way is to end with RAA, under the premise φφ. Using that
premise φφ and the starting premise
¬(φ∨ψ)¬(φ∨ψ) can you derive the contradiction?
| 1 | ¬(φ∨ψ)¬(φ∨ψ) | | Premise
|
| 2 | subproof:φ ⊢ falseφ ⊢ | | |
| 2.a | | φφ | Premise for subproof
|
| 2.b | | (φ∨ψ)(φ∨ψ) | ∨∨Intro, line 2a
|
| 2.c | | false | falseIntro, lines 1,2b
|
| 3 | ¬φ¬φ | | RAA, line 2
|
Using the inference rule RAA, prove
¬φ ⊢ ¬(φ∧ψ)¬φ ⊢ ¬(φ∧ψ).
Show that
(¬W-safe∨¬Y-unsafe) ⊢ (W-unsafe∨Y-safe)(¬W-safe∨¬Y-unsafe) ⊢ (W-unsafe∨Y-safe).
The proof is a bit longer than you might expect.
Use the ∨∨Elim inference rule to get the final result.
In our inference rules (unlike our equivalences),
we have nothing corresponding to DeMorgan's Law.
Prove each of the following versions.
-
(φ∨ψ) ⊢ ¬(¬φ∧¬ψ)(φ∨ψ) ⊢ ¬(¬φ∧¬ψ)
-
¬(φ∨ψ) ⊢ (¬φ∧¬ψ)¬(φ∨ψ) ⊢ (¬φ∧¬ψ)
-
(φ∧ψ) ⊢ ¬(¬φ∨¬ψ)(φ∧ψ) ⊢ ¬(¬φ∨¬ψ)
-
¬(φ∧ψ) ⊢ (¬φ∨¬ψ)¬(φ∧ψ) ⊢ (¬φ∨¬ψ)
The
above exercise suggests that it would be useful to have
an inference rule or theorem that says given
θ ⊢ ¬δθ ⊢ ¬δ,
then
¬θ ⊢ δ¬θ ⊢ δ.
Or, equivalently, because of →→Intro and →→Elim,
(θ→¬δ) ⊢ (¬θ→δ)(θ→¬δ) ⊢ (¬θ→δ).
Why don't we?
In our inference rules (unlike our equivalences),
we have nothing that directly equates
(φ→ψ)(φ→ψ) and (¬φ∨ψ)(¬φ∨ψ).
Prove each of the following.
-
(φ→ψ) ⊢ (¬φ∨ψ)(φ→ψ) ⊢ (¬φ∨ψ)
-
(¬φ∨ψ) ⊢ (φ→ψ)(¬φ∨ψ) ⊢ (φ→ψ)
Prove the following:
(φ→ψ), (ψ→φ) ⊢ ((φ∧ψ)∨(¬φ∧¬ψ))(φ→ψ), (ψ→φ) ⊢ ((φ∧ψ)∨(¬φ∧¬ψ))
Prove what is commonly called the
Law of Excluded Middle:
⊢ (χ∨¬χ)⊢ (χ∨¬χ).
-
Give a short proof citing our
previous proof of
⊢ ¬ (χ∧¬χ) ⊢ ¬ (χ∧¬χ)
and the relevant version of DeMorgan's Law from
above.
-
Give a direct version without using previous theorems.
Use RAA two or three times.
Prove the missing steps and reasons in
the following WaterWorld proof of
X-has-1 ⊢ (W-unsafe∨Y-unsafe)X-has-1 ⊢ (W-unsafe∨Y-unsafe).
| 1 | X-has-1X-has-1 | |
____________
|
| 2 | ____________ | |
WaterWorld domain axiom
|
| 3 | ____________ | |
→→Elim, lines 1,2
|
| 4 | subproof:(W-safe∧Y-unsafe) ⊢ (W-unsafe∨Y-unsafe)(W-safe∧Y-unsafe) ⊢ (W-unsafe∨Y-unsafe) | | |
| 4.a | | (W-safe∧Y-unsafe)(W-safe∧Y-unsafe) |
Premise for subproof
|
| 4.b | | Y-unsafeY-unsafe |
____________
|
| 4.c | | (W-unsafe∨Y-unsafe)(W-unsafe∨Y-unsafe) |
____________
|
| 5 | subproof:¬(W-safe∧Y-unsafe) ⊢ (W-unsafe∨Y-unsafe)¬(W-safe∧Y-unsafe) ⊢ (W-unsafe∨Y-unsafe) | | |
| 5.a | | ¬(W-safe∧Y-unsafe)¬(W-safe∧Y-unsafe) |
Premise for subproof
|
| 5.b | | (W-unsafe∧Y-safe)(W-unsafe∧Y-safe) |
CaseElim (left), lines ____________,
where (φ=____________)(φ ____________),
and (ψ=____________)(ψ ____________)
|
| 5.c | | ____________ |
____________
|
| 5.d | | (W-unsafe∨Y-unsafe)(W-unsafe∨Y-unsafe) |
____________
|
| 6 | ((W-safe∧Y-unsafe)∨¬(W-safe∧Y-unsafe))((W-safe∧Y-unsafe)∨¬(W-safe∧Y-unsafe)) | |
Theorem: Excluded Middle,
where (χ=____________)(χ ____________)
|
| 7 | (W-unsafe∨Y-unsafe)(W-unsafe∨Y-unsafe) | |
____________
|
[Practice problem—solution provided.]
Given the
above figure, and using any of the immediately obvious facts
as premises, prove that location PP is safe
by using our proof system and the WaterWorld domain axioms.
While this proof is longer (over two dozen steps),
it's not too bad when sub-proofs are used appropriately.
To make life easier,
you may use the following theorem:
(Q-has-1→((P-safe∧R-safe)∨(P-safe∧W-safe)∨(R-safe∧W-safe)))(Q-has-1→((P-safe∧R-safe)∨(P-safe∧W-safe)∨(R-safe∧W-safe))),
along with any proven previously.
When looking at the given board, you can use premises like
Y-safeY-safe as well as ¬Y-unsafe¬Y-unsafe.
| 1 | Q-has-1Q-has-1 | |
Premise
|
| 2 | X-has-1X-has-1 | |
Premise
|
| 3 | ¬Y-unsafe¬Y-unsafe | |
Premise
|
| 4 | (W-unsafe∨Y-unsafe)(W-unsafe∨Y-unsafe) | |
Theorem:
above problem, line 2
|
| 5 | (Y-unsafe∨W-unsafe)(Y-unsafe∨W-unsafe) | |
Theorem: ∨∨ commutes, line 4
|
| 6 | W-unsafeW-unsafe | |
CaseElim, lines 3,5
|
| 7 | subproof:¬¬(P-safe∧W-safe) ⊢ false¬¬(P-safe∧W-safe) ⊢ | | |
| 7.a | | ¬¬(P-safe∧W-safe)¬¬(P-safe∧W-safe) | Premise for subproof
|
| 7.b | | (P-safe∧W-safe)(P-safe∧W-safe) | ¬¬Elim,
line 7.a
|
| 7.c | | W-safeW-safe |
∧∧Elim, line 7.b
|
| 7.d | | (W-safe→¬W-unsafe)(W-safe→¬W-unsafe) |
WaterWorld domain axiom
|
| 7.e | | ¬W-unsafe¬W-unsafe |
→→Elim, lines 7.c,7.d
|
| 7.f | | false |
falseIntro, lines 6,7.e
|
| 8 | ¬(P-safe∧W-safe)¬(P-safe∧W-safe) | |
RAA, line 7
|
| 9 | subproof:¬¬(R-safe∧W-safe) ⊢ false¬¬(R-safe∧W-safe) ⊢ | | |
| 9.a | | ¬¬(R-safe∧W-safe)¬¬(R-safe∧W-safe) | Premise for subproof
|
| 9.b | | (R-safe∧W-safe)(R-safe∧W-safe) |
¬¬Elim, line 9.a
|
| 9.c | | W-safeW-safe |
∧∧Elim, line 9.b
|
| 9.d | | (W-safe→¬W-unsafe)(W-safe→¬W-unsafe) |
WaterWorld domain axiom
|
| 9.e | | ¬W-unsafe¬W-unsafe |
→→Elim, lines 9.c,9.d
|
| 9.f | | false |
falseIntro, lines 6,9.e
|
| 10 | ¬(R-safe∧W-safe)¬(R-safe∧W-safe) | |
RAA, line 9
|
| 11 | (Q-has-1→(((P-safe∧R-safe)∨(P-safe∧W-safe))∨(R-safe∧W-safe)))(Q-has-1→(((P-safe∧R-safe)∨(P-safe∧W-safe))∨(R-safe∧W-safe))) | |
Theorem: Allowed by problem statement
|
| 12 | (((P-safe∧R-safe)∨(P-safe∧W-safe))∨(R-safe∧W-safe))(((P-safe∧R-safe)∨(P-safe∧W-safe))∨(R-safe∧W-safe)) | |
→→Elim, lines 1,11
|
| 13 | ((R-safe∧W-safe)∨((P-safe∧R-safe)∨(P-safe∧W-safe)))((R-safe∧W-safe)∨((P-safe∧R-safe)∨(P-safe∧W-safe))) | |
Theorem: ∨∨ commutes, line 12
|
| 14 | ((P-safe∧R-safe)∨(P-safe∧W-safe))((P-safe∧R-safe)∨(P-safe∧W-safe)) | | CaseElim, lines 8,13
|
| 15 | ((P-safe∧W-safe)∨(P-safe∧R-safe))((P-safe∧W-safe)∨(P-safe∧R-safe)) | | Theorem: ∨∨ commutes, line 14
|
| 16 | (P-safe∧R-safe)(P-safe∧R-safe) | |
CaseElim, lines 10,15
|
| 17 | P-safeP-safe | |
∧∧Elim, line 16 — Whew!
|
Alternatively, the subproofs could easily have been pulled out into
lemmas. Just like using subroutines in a program, that would
make the proof somewhat clearer, even though in this case
each lemma would be used only once.
Observe how the two subproofs have some identical lines
(7.c-7.f and 9.c-9.f).
It would be incorrect to replace those lines in the second subproof
with a citation of the results of the first subproof —
first, because the previous subproof had been completed, and
moreover, the two subproofs have different premises.
This is analogous to two subroutines that happen to have some
identical code lines, even through they are called separately and have
different parameters.
Interestingly, we didn't need to use R-safeR-safe as a premise.
(In fact, we nearly proved that
¬R-safe¬R-safe would have been inconsistent
with the other premises.)
Starting from the WaterWorld axiom
(Q-has-1→((P-safe∧R-safe∧W-unsafe)∨(P-safe∧R-unsafe∧W-safe)∨(P-unsafe∧R-safe∧W-safe)))(Q-has-1→((P-safe∧R-safe∧W-unsafe)∨(P-safe∧R-unsafe∧W-safe)∨(P-unsafe∧R-safe∧W-safe))),
we could prove the following theorem cited in the
previous problem:
(Q-has-1→((P-safe∧R-safe)∨(P-safe∧W-safe)∨(R-safe∧W-safe)))(Q-has-1→((P-safe∧R-safe)∨(P-safe∧W-safe)∨(R-safe∧W-safe))).
Prove the following theorem which is slightly simpler:
(φ→((ψ∧θ)∨(δ∧κ))) ⊢ (φ→(ψ∨δ))(φ→((ψ∧θ)∨(δ∧κ))) ⊢ (φ→(ψ∨δ)).
If you have trouble, first prove an even simpler version:
(φ→(ψ∧θ)) ⊢ (φ→ψ)(φ→(ψ∧θ)) ⊢ (φ→ψ).
[Practice problem—solution provided.]
Show that the ¬¬Elim inference rule is redundant in our system.
In other words, without using ¬¬Elim, prove that
¬¬φ ⊢ φ¬¬φ ⊢ φ.
| 1 | ¬¬φ¬¬φ | |
Premise
|
| 2 | subproof:¬φ ⊢ false¬φ ⊢ | | |
| 2.a | | ¬φ¬φ |
Premise for subproof
|
| 2.b | | false |
falseIntro, lines 1,2.a
|
| 3 | φφ | |
RAA, line 2
|
Show that the ¬¬Intro inference rule is redundant in our system.
In other words, without using ¬¬Intro, prove that
φ ⊢ ¬¬φφ ⊢ ¬¬φ.
Show that the CaseElim inference rule is redundant in our system.
For brevity, we'll just consider the left-hand version.
In other words, without using CaseElim, prove that
(φ∨ψ), ¬φ ⊢ ψ(φ∨ψ), ¬φ