Shave and a Haircut!
First there was …
Then came along …
The following explanation comes from one of my absolute favorite college textbooks (and personal math bibles) Discrete Mathematics with Applications by Dr. Susanna S. Epp.
For the most part, sets are not elements of themselves. A perfect example would be to say that the set of all cats is not a cat, which helps to explain that the set of all integers is not an integer. However, in some cases, sets are elements of themselves. So, to explain, the set of all salt particles is salt particles, which helps to explain that the set of all abstract ideas is an abstract idea.
So, to explain Russell’s paradox
S=\lbrace N |\space N \space \text {is a set and} \space N \notin N \rbrace
The question then, is S an element of itself? In other words, Does the barber shave himself?
The answer is Neither yes nor no. If the barber shaves himself, then he’s a member of the group of men who shave themselves.
\text {If} \space S\in S, \text {then} \space S \space \text {satifies the defining property for} \space S, \text{therefore} \space S\notin S.
\text {But if} \space S\notin S, \text {then} \space S \space \text {is a set such that} \space S\notin S,
\text {and} \space S \space \text {satisfies the defining property for} \space S.
\text {This implies that} \space S\in S.
\text{Therefore, neither is} \space S\in S, \text {nor is} \space S\notin S.
This is a contradiction.
Russell’s paradox, however, is no longer contradictory when we use a predicate as a defining property and specify that the set is a subset of a known set.
By doing so, we remove the property “the set of all sets that are not elements of themselves.” As a result, we only address the property that “the set of all sets that are subsets of some known set and that are not elements of themselves.”
Now it’s not a contradiction.
\text {Let} \space L \space \text {be a set of sets and let} \space S=\lbrace N \in L|N \notin N \rbrace. \space \text {Is} \space S\in S?
No, it is not.
\text {If} \space S \in S, \text {then} \space S \space \text {satisfies the defining property for} \space S.
\text {Therefore,} \space S \notin S.
\text{When the definition of} \space S \space \text {is changed, only}
S \in S \space --> \space S \notin S \space \text{can be proved.}
\text {Therefore} \space S\notin L.
…which is not contradictory.
But THEN along came.….
…who showed us all that it is not possible to prove in a mathematically rigorous way that mathematics is free of contradictions!