Law of Noncontradiction (LNC): ¬(P∧¬P)
Law of Excluded Middle (LEM): P∨¬P
If you apply the Morgan's Laws, you can see that:
¬(P∧¬P) ⟺ (¬P ∨ P)
Hence, they seemingly express one and the same law ... but they don't.
The problem is that the proof for De Morgan's laws uses truth tables. The use of truth tables implies the LEM already. So, it is indeed circular.ChatGPT: Why does the Law Of the Excluded Middle not follow from the law of noncontradiction using the Morgan's law?
(Remark: Aka, what's wrong with the proof?)
Can LEM Follow from LNC via De Morgan's Laws?
To derive P∨¬P (LEM) from ¬(P∧¬P) (LNC), let's consider the logic:
Start with ¬(P∧¬P), the statement of LNC.
Using De Morgan's laws, ¬(P∧¬P)≡¬P∨¬(¬P)
Simplify ¬(¬P) to P, giving:
¬(P∧¬P)≡¬P∨P.
Rearrange the disjunction:
¬P∨P≡P∨¬P.
At this stage, it seems we've derived LEM from LNC. However, this reasoning assumes that the rules of classical logic, such as De Morgan's laws and the double negation elimination (¬(¬P)≡P¬(¬P)≡P), are valid.
Why This Is Circular Reasoning
The validity of De Morgan's laws and double negation elimination depends on classical logic, which assumes the Law of Excluded Middle as a foundational principle. In other words:
To use De Morgan's laws and double negation elimination, you're already working within a framework where LEM is valid.
Therefore, deriving P∨¬P (LEM) from ¬(P∧¬P) (LNC) using De Morgan's laws is circular.
Is this a Gettier problem?
So, the expression is indeed technically true:
¬(P∧¬P) ⟺ (¬P ∨ P)
The expression is also indeed technically provable, i.e. perfectly justified.
So, it is a "justified true belief" (JTB) that the LEM necessarily follows from the LNC, but it is actually false.
So, is this a Gettier problem?