I've always thought the phrase "the exception that proves the rule" is nonsense, btw. There's no such "proof."

I think the intent behind the modern usage of "the exception that proves the rule" is that:

a) the rule is a heuristic, not a hard-and-fast rule, and

b) the notability of exceptions suggests that the heuristic is pretty good.

In other words, if the exception is, in fact, exceptional, it reinforces that the heuristic is reasonably valid for non-exceptional cases.

In this case, the fact that it took the intervention of the family-friend-who-works-at-Google-HQ to resolve a pay dispute is an exceptional circumstance, and reinforces that for most people, because presumably most people don't have a family friend who works at Google HQ, there is no recourse to pay disputes.

Hence, the exception—by being exceptional—proves the heuristic rule that most employees of these contractor corps have little power to negotiate for their rights.

And the "proof" used is an archaic term that means "test". The same word occurs in discussing alcohol strength ("100 proof") and in "proof standard" (the standard[1] that you use to compare against).

The German cognate would be "prüfen", the Swedish probably "pröva".

So, the exception tests the rule.

[1] And the standard is, of course, a reference exemplar of some sort.

Yeah, or for example in Dutch 'scheikundeproef' or 'natuurkundeproef' which we did at high school: a test involving the laws of chemistry or physics respectively.

That’s a great explanation.

Another commonly misused phrase is "begging the question" which you have done here. It means "assuming that which you need to prove."

You have no proof that it was exceptional. All you have is one case. You have no idea what other people in pay disputes end up doing.

First off, I was just framing the parent’s post in the modern usage of “exception that proves the rule” and not making any claims myself, so perhaps reconsider who’s making unfounded assumptions here.

> You have no idea what other people in pay disputes end up doing.

But also this is a super weak argument. I don’t have “no idea”—I have an educated guess that most people don’t have family friends who work at Google. Most people don’t have the resources to sue an employer over a few hours of unpaid labor. Wage theft is one of the most extensive unprosecuted crimes, that’s been studied and shown many times.

Are you suggesting that one should assume that most low-level employees in pay disputes get happy outcomes? That seems…naive.

The question people are asking is "How would your relative have gotten paid if you hadn't worked at Google"?

This kind of pedantry about which words people use to make an argument is literally garbage.

To continue the spirit of the many nested comments now going back and forth:

Metaphorically.

People who have high regard for it are literally queer. Reference: myself.

What's 'it' here, I can't tell if you have high regard for pedantry about words or if you have high regard for semantic drift.

It's beautiful and full of stars.

The latter

I appreciate the compliment.

I think you mean figuratively garbage. Words can only be literally garbage if they're printed on a page which is then disposed of.

I mean, while we're being pedantic we may as well get it right.

People are allowed to use a phrase for its literal meaning instead of the traditional idiomatic meaning.

"beg the question" does not have a literal meaning. It makes no sense unless you're so used to hearing it used that way.

"beg" often means "beg for". "This situation begs for the question:" is a completely reasonable way to start a sentence.

Or to go with synonyms. Solicits the question. Entreats the question. These are the literal meanings of the words, and they work fine.

It is an old phrase. In this case, "proves" means something more like "tests". In modern terms, I'd say it's something like, "You don't really know you have a rule until someone has tested it by claiming an exception." e.g., setting a child's bedtime to 9pm will pose no challenge if they voluntarily go to bed at 8:30, it isn't until the exception comes along that you know whether or not you really have a rule.

I think the phrase 'proofing yeast' in bread making has the same connotations - you're testing that the yeast is still alive.

Off-topic here, but since I'm often making pizza dough:

I always wondered about that word. If you want to test yeast, you just put it in some sugar water & see if it foams.

But then, I guess you can't do that if it's mixed with flour & other stuff?

The word "prove" used to also encompass the meaning of "challenge" or "put to the test".

It still retains that archaic meaning in the phrase you mentioned and terms like "proving grounds".

Edit:

And like so many other proverbs like "A few bad apples", "Blood is thicker than water", "Carpe diem", "The exception proves the rule" means exactly the opposite of what most people think it does.

In the case of “blood is thicker than water,” unfortunately, the more historically common understanding is also its most common understanding today, and has been for so many centuries that we dont actually know if it originated with its original meaning or it’s inverse.

I’m bringing this up because, for many years, I was one of the folk who spread the idea that it actually was supposed to mean “the blood of the covenant is thicker than the water of the womb.” But there’s no strong historical basis for this, lovely as the idea may be.

—

As for the “bad apple” one, I didn’t realize that it was commonly misunderstood. I always felt like someone mentioning that someone was “a bad apple” was an almost violent accusation: the person needed to be extricated from the group, less the group become infected by their corruption. I am now trying to think back to times when I’ve heard this one used and wondering if I misunderstood the intent.

> The word "prove" used to also encompass the meaning of "challenge" or "put to the test".

Yes but "There is little evidence of the phrase being used in this second way.", citations located at wikipedia.

So I think it makes more sense to go with the interpretation of "if someone lists an exception, then that's proof of a more general rule that it's an exception from". But with that use, something specifically has to be an exception. Anyone applying "exception proves the rule" to an anecdote is using it wrong.

Also "proof reading" and "100 proof whisky".

And proofing dough.

The exception that proves the rule is from a legal argument by Cicero.

It means that if someone bothers to write an exception in a law, say e.g. "but in case window is broken by stone smaller than 1 cm in diameter, then it's the owner of the window that should pay reparations", then in the general case it's likely the thrower of the stone that pays, otherwise there would have been no point in adding so specific an exception.

This is the correct answer. "Exception proves the rule" = a documented exception is an evidence that a general rule exists.

When you see text "kids under age of 3 years don't have to pay", you can conclude that older people are supposed to pay.

The analogy for Google situation would be if hypothetically someone found an internal e-mail written by the boss of Google HR department saying "contractors who have relatives in Google, or who have published their story on Hacker News front page, need to be fully paid for their work". Then we could conclude that other contractors typically don't get fully paid, even if there is nothing in the e-mail saying that.

It's so misused most of the time. The clearest, correct example of it I've seen is a sign that says "No Parking from 10PM to 5AM". This is an exception to the otherwise unstated rule that parking is allowed during all other times, therefore proving it exists.

I disagree with most of the commenters that want to apply the phrase to heuristic claims.

This is currently the only reply that's downvoted below zero, but this is also the only reply that matches the definition of the phrase as I understood it.

The phrase is misunderstood. Historically "Proof" is contextually meaning the nature of the test, not passing it. "failed the test"-proof has been "disproved" by this exception, which shows a belief does not generally apply. I'm not explaining this very well but a Google search will throw up better explanations.

Some how it's morphed to colloquially mean "yes, that one exception aside, I'm right and you're wrong" as if the negative somehow reinforces the positive..

The phrase "the proof of the pudding is in the eating" helps understand this. The pudding may prove to be bad or good. Find which proof by eating it.

jerf's explanation makes sense. The negative certainly does not reinforce the positive.

Yes. They said it better. It certainly doesn't reinforce the positive, yet people seem to insist on using it as if it does. "Well, yes, that penny did land on it's edge but the exception proves the rule it had to be either heads or tails" type statements abound. No.. the exception proved pennies have three states not two: your argument is busted.

Exception proves the rule comes from Roman law: if there is a sign saying you can’t park here on Saturdays, it means that you can park here on any other day.

I was taught it comes from logic, and argumentation: the act of stating an exception proves a rule as stated (a rule is a claim to an axiom) is wrong. So it's in logic and philosophy, not Roman law.

If you’re reading a guide book and it specifically says: “Use spell X to ability kill this boss” the implication is you can’t instantly kill other bosses by using ability X.

Logically the statement says nothing beyond what it says, but implicitly by providing the exception you’re stating that the general rule is different.

It’s a perfectly rational linguistic argument, though often misused outside of that context.

The "rule" is the null hypothesis. The exception is a result about three standard deviations from the null hypothesis. Finding such an exception result means you can not reject the null hypothesis.

The argument is that not being able to reject the null hypothesis means that the null hypothesis is correct. This is an invalid argument. We're sorry but we can not accept your submission at this time.

Agreed. It is infuriating because it is internally inconsistent on its face. The way logic works, an exception disproves the rule. Surprise, surprise.

No. It is a legal principle. If there is a sign stating: "Parking prohibited 03am - 07am" then that implies that parking is allowed outside that window of time.

I can summarize OPs post even more succinctly: Nepotism in google only way to get paid.