Laws of correct reasoning; sophisms and paradoxes
Every argument has several statements, the premises, and a conclusion drawn from them, joined by “therefore”. If the premises are S and the conclusion is R, the argument has the form S ⇒ R and is correct if this expression is a tautology. Four basic correct forms are ((p ⇒ q) ∧ p) ⇒ q; ((p ⇒ q) ∧ ¬q) ⇒ ¬p; ((p ∨ q) ∧ ¬p) ⇒ q; ((p ⇒ q) ∧ (q ⇒ r)) ⇒ (p ⇒ r). But ((p ⇒ q) ∧ q) ⇒ p is not a tautology (it is false when p is false and q is true), so such a conclusion is wrong. A sophism is a false “proof” whose mistake is cleverly hidden (often a hidden division by zero). A paradox is an unexpected statement that contradicts an accepted view; for example “This sentence is false” leads to a contradiction whether we call it true or false.
| Name | Form |
|---|---|
| modus ponens | ((p ⇒ q) ∧ p) ⇒ q |
| modus tollens | ((p ⇒ q) ∧ ¬q) ⇒ ¬p |
| disjunctive syllogism | ((p ∨ q) ∧ ¬p) ⇒ q |
| chain rule | ((p ⇒ q) ∧ (q ⇒ r)) ⇒ (p ⇒ r) |
A chat message says: “Every official notice contains a link. This message contains a link, so it is official.” The conclusion is wrong, because other messages also contain links. The correct direction is the reverse: a message without a link is not one of those official notices.
Fact-checkers put the claims of rumours and adverts into logical form, judges avoid errors when drawing conclusions from evidence, and programmers reason this way when hunting for a bug.
On a notice or advert at school find a claim of the form “if A, then B”. Write its logical form and decide whether the conclusion is valid.
| p | q | p ⇒ q | (p ⇒ q) ∧ q | ((p ⇒ q) ∧ q) ⇒ p |
|---|---|---|---|---|
| T | T | T | T | T |
| T | F | F | F | T |
| F | T | T | T | F |
| F | F | T | F | T |
“Find the mistake” contest: each group gets a simple sophism prepared by the teacher (for example the “proof” of 2 = 1); the group finds the step with a division by zero or a wrong conclusion and proves it.