Showing posts with label Pierce's Lemma. Show all posts
Showing posts with label Pierce's Lemma. Show all posts

Thursday, 10 April 2008

The Proof is in the P(udding)ierce's Lemma

Hey there I just unearthed an old proof that I liked from my degree. I figured some folk may appreciate it so here it is.

If you don't appreciate it then please enjoy the cartoon above that I borrowed from here.

Pierce's Lemma

Show that: ¦-- (((a->b)->a)->a) (=: c)

Pf: So rewrite the lemma as ¦-- c
By one of the "famous" theorems of prop. calc. we have

¦-- ((a->c)->((¬a->c)->c))

So Suff. To Prove (i)¦--(a->c) & (ii)¦--(¬a->c)

(i) by applying DT twice it is STP that

{a, ((a->b)->a)} ¦-- a

a one line pf so done

(ii) by applying DT twice it is STP that

{¬a, ((a->b)->a)} ¦-- a

One of the theorems says: ¦-- (¬a->(a->b))
So we have that {¬a, ((a->b)->a)} ¦-- ¬a
{¬a, ((a->b)->a)} ¦-- (¬a->(a->b))
so by MP: {¬a, ((a->b)->a)} ¦-- (a->b)
{¬a, ((a->b)->a)} ¦-- ((a->b)->a)
so by MP: {¬a, ((a->b)->a)} ¦-- a

So we've proved (i) and (ii) and so by the tautology we have Peirce's Lemma.