"There is an Exception to every rule"

[lorb]

Is the sentence "There is an Exception to every rule" a version of the liars paradox? Because for it to be true, it must itself have an exception, but that would make it false. Is this the same as the liars paradox, or something different?

[Eebster the Great]

It is not paradoxical, it is simply false. There are rules which have no exceptions.

[stopmadnessnow]

If it is a physical law, would it necessarily be the same just after the Big Bang?

[Soupspoon]

"There is an exception to every rule" is not itself a Rule, it is an Aphorism. Rules, themselves, are not Laws, and Laws are not Truths, which are in turn different from Facts that may or may not be backed up with Data. This post is a Summary, and as such contains Innacuracies and Assumptions for the sake of Brevity. Including any claim that it exhibits anything other than Verbose Waffling largely uninfluenced by Analysis of Facts for the possible Purposefulness of conveying Humour.

Hope That Helps. Have A Nice Day.

[ConMan]

Epimenides of Crete is claimed to have said "All Cretans are liars". If true, then it means that all Cretans are liars, but Epimenides being a Cretan himself must then be a liar, meaning his statement cannot be true. If the statement is false, then it means that Epimenides was lying, and that not all Cretans are liars. It is still consistent that Epimenides is a liar, as long as there is at least one Cretan who tells the truth.

Similarly, if we assume that "There is an exception to every rule" is meant to be a rule, then if it's true it must have an exception, and if that's the case then there is a rule without an exception and hence the original rule is false. So therefore "There is not an exception to every rule" is the true rule, and there must be at least one rule that has no exception.

tl;dr The negation of "For all X, Y" is "There exists X such that not Y", not "For all X, not Y".

[doogly]

Similarly, if we assume that "There is an exception to every rule" is meant to be a rule, then if it's true it must have an exception, and if that's the case then there is a rule without an exception and hence the original rule is false.

"There is an exception to every rule" can be true AND have an exception -- it is a rule, A, about rules. If the exception to A is A itself, then we're fine!

Though most people take the definition of a rule to be a thing with no exceptions, so it would just be a stupid nonmathematical statement rather than a rule at all, really.

[xkcdfan]

"Every even number is divisible by 2 with no remainder" - there's your exception to "There's an exception to every rule".

[Zohar]

That's not a rule, that's a tautology.

[DavCrav]

That's not a rule, that's a tautology.

Fine, for every prime p, and every integer a prime to p, a raised to power p minus a is divisible by p. Not a tautology, definitely no exceptions. Happy now?

[Eebster the Great]

That's not a rule, that's a tautology.

Fine, for every prime p, and every integer a prime to p, a raised to power p minus a is divisible by p. Not a tautology, definitely no exceptions. Happy now?

If you mean "For every prime p and integer a coprime to p, (a^p - a) | p," then that is also a tautology.

[WibblyWobbly]

That's not a rule, that's a tautology.

Fine, for every prime p, and every integer a prime to p, a raised to power p minus a is divisible by p. Not a tautology, definitely no exceptions. Happy now?

If you mean "For every prime p and integer a coprime to p, (a^p - a) | p," then that is also a tautology.

Pardon my stupidity, but why is that a tautology? Does it not have to be proven?

[Flumble]

That's not a rule, that's a tautology.

Fine, for every prime p, and every integer a prime to p, a raised to power p minus a is divisible by p. Not a tautology, definitely no exceptions. Happy now?

If you mean "For every prime p and integer a coprime to p, (a^p - a) | p," then that is also a tautology.

To be perfectly pedantic, this seems to be a logical validity, not a tautology.


I think a nice rule without exception is "I think", in as much as it can be considered a rule. ("you may not kill someone" is considered a rule, so every "<entity> <action>" can be considered a rule, right?)
No one* having that thought can deny they* think and are an entity.

*for lack of a better word.

[Eebster the Great]

That's not a rule, that's a tautology.

Fine, for every prime p, and every integer a prime to p, a raised to power p minus a is divisible by p. Not a tautology, definitely no exceptions. Happy now?

If you mean "For every prime p and integer a coprime to p, (a^p - a) | p," then that is also a tautology.

To be perfectly pedantic, this seems to be a logical validity, not a tautology.

I guess so. The tautology would be "(p is a prime) ∧ (a is an integer coprime to p) → ((a^p - a) | p)."

[dalcde]

That's not a rule, that's a tautology.

Fine, for every prime p, and every integer a prime to p, a raised to power p minus a is divisible by p. Not a tautology, definitely no exceptions. Happy now?

If you mean "For every prime p and integer a coprime to p, (a^p - a) | p," then that is also a tautology.

To be perfectly pedantic, this seems to be a logical validity, not a tautology.

I guess so. The tautology would be "(p is a prime) ∧ (a is an integer coprime to p) → ((a^p - a) | p)."

No. That's a falsehood.

[rmsgrey]

If a is not coprime to p, then it's a multiple of p, so a^p and a are also multiples of p, and so is their difference.

[Meteoric]

The difference being a multiple of p would be denoted p|(a^p-a), though.

[WibblyWobbly]

If a is not coprime to p, then it's a multiple of p, so a^p and a are also multiples of p, and so is their difference.

Forgive me, but isn't 6 not coprime to 9? They have a factor in common, but neither is a multiple of the other.

[rmsgrey]

If a is not coprime to p, then it's a multiple of p, so a^p and a are also multiples of p, and so is their difference.

Forgive me, but isn't 6 not coprime to 9? They have a factor in common, but neither is a multiple of the other.

Last time I checked, 9 isn't a prime...

[WibblyWobbly]

If a is not coprime to p, then it's a multiple of p, so a^p and a are also multiples of p, and so is their difference.

Forgive me, but isn't 6 not coprime to 9? They have a factor in common, but neither is a multiple of the other.

Last time I checked, 9 isn't a prime...

Well maybe you should check again. Or maybe I should stop posting in this thread.

[MartianInvader]

9 is a prime example of a square number.

[Eebster the Great]

And a squared example of a prime number.

[MartianInvader]

This got me thinking - 18 is a really boring, uncool number. Like, unusually so, so much that nothing could be more boring or uncool. I'd definitely choose 18 as an ususual example of a number of unparalleled uncoolness.

In other words, 18 is a prime odd perfect square.

[PM 2Ring]

This got me thinking - 18 is a really boring, uncool number. Like, unusually so, so much that nothing could be more boring or uncool. I'd definitely choose 18 as an ususual example of a number of unparalleled uncoolness.

In other words, 18 is a prime odd perfect square.

Here's a reasonably cool property of 18.

18² + 1 = 325 = 13 x 5², so 18 / 5 is the best approximation of sqrt(13) in that neighbourhood. Note that 13 = 2² + 3², so sqrt(13) is the length of the diagonal of the smallest rectangle with unequal (integer) sides > 1.

It's an interesting "coincidence" that 18 - 5 = 13 as well as 18 / 5 ≈13^½.

My favourite rule about primes:

"All primes are odd, except 2, which is the oddest of them all." — Graham, Knuth, Patashnik, Concrete Mathematics, second edition, page 129.

[jewish_scientist]

Epimenides of Crete is claimed to have said "All Cretans are liars". If true, then it means that all Cretans are liars, but Epimenides being a Cretan himself must then be a liar, meaning his statement cannot be true. If the statement is false, then it means that Epimenides was lying, and that not all Cretans are liars. It is still consistent that Epimenides is a liar, as long as there is at least one Cretan who tells the truth.

Similarly, if we assume that "There is an exception to every rule" is meant to be a rule, then if it's true it must have an exception, and if that's the case then there is a rule without an exception and hence the original rule is false. So therefore "There is not an exception to every rule" is the true rule, and there must be at least one rule that has no exception.

tl;dr The negation of "For all X, Y" is "There exists X such that not Y", not "For all X, not Y".

The Oxford Dictionary defines 'liar' as "A person who tells lies." In other words, there is a non-zero probability that a statement by a liar is false. This probability does not have to be one, so a liar can make a say a statement that is true. Therefor, Epimenides of Crete's statement can be true without creating a paradox.

[Soupspoon]

If it were "are compulsive liars", it may be a problem, if that assures the required absolute.

Or, then again, Epimenides of Crete may have assumed that being told that he was Cretan is a lie, by his habitually lying parents, and his statement is a perceived truth but, due to his misapprehension of the truth (that he and/or his parents actually did not (think they made a) lie) he is effectively lying anyway.

Or Epimenides of Crete actually was from Athens. Those habitually lying Athenians, right?!?

[Eebster the Great]

I think it is pretty clear that while Epimenides was trying to get across a genuinely interesting paradox, he did a spectacularly poor job of it.

[ConMan]

So if Epimenides had said "All Cretans are poor expressers of logical statements" ...

[Flumble]

Perhaps Epimenides did formulate the paradox correctly, but it was lost in translation: we're speaking of attributions to a semi-mythical[1] person in a dialect used 2 millenia ago.

[Soupspoon]

Don't be silly. If it is written down, regardless of how many intermediate translations were made, it is True.

(Unless it was ever written by a Cretan, obviously!)

[Flumble]

That rule only applies to the internet!
...which makes for a interesting paradox now that news agencies publish the same articles on the internet and in the newspaper itself, which we know is full of lies.

[Eebster the Great]

That rule only applies to the internet!
...which makes for a interesting paradox now that news agencies publish the same articles on the internet and in the newspaper itself, which we know is full of lies.

Mainstream Media is all Fake Lies though. You can only get real lies on alt media like Breitbart.

[Soupspoon]

(Not really germaine to the mathematical logic, but on UK keyboards, the right Alt key is "Alt Gr", thus confirming that everything under the banner of Alt-Right was always supposed to be aggressive and threatening... Grrrrr!)

[Xanthir]

18² + 1 = 325 = 13 x 5², so 18 / 5 is the best approximation of sqrt(13) in that neighbourhood. Note that 13 = 2² + 3², so sqrt(13) is the length of the diagonal of the smallest rectangle with unequal (integer) sides > 1.

It's an interesting "coincidence" that 18 - 5 = 13 as well as 18 / 5 ≈13^½.

18 - 5 = 13
18 / 5 ≈ √13
18 + 5 = 23
2 + 3 = 5

MATH ILLUMINATI CONFIRMED