Talk:Exception that proves the rule

--CopyToWiktionaryBot 11:32, 1 February 2007 (UTC)Reply

It would be good in this case to avoid the POV terms "correct" and "incorrect" in reference to this expression. After all, the "incorrect" usage goes back to the 17th century, and has established itself in its own right. The reader who encounters the expression needs to be aware of both possibilities. In no way should he assume that the more subtle older meaning is used. There are many more usages in English where a more modern change has been has been accepted in a much quicker time.

The other point is that "proves" is not equivalent to "implies" or even "infers". Let's not forget the in casibus non exceptis in the Latin expression. In the scientific method there is a repeating cycle of hypotheses and proofs. As long as we are at the hypothesis stage we can only state that there may be a rule; with that information alone we cannot be sure of what it is. It's also important that the exception be seen as an exception, and that the supposed rule be unique and unambiguous. If it can be an exception to more that one conflicting rule it is no longer effective as a proof. What makes your parking example interesting is the you phrased it as parking allowed on Sunday whereas the source which uses this has the parking prohibited on Sunday. In a common law tradition anything that is not specifically prohibited is allowed. Thus it would be reasonable to say that if parking is prohibited on Sunday it must be allowed on other days. It would not be reasonable to infer that because parking is allowed on Sunday it must be prohibited on the other days. This expression is far more subtle than it has been made out to be. Eclecticology 07:37, 3 November 2007 (UTC)Reply

Re: "It would be good in this case to avoid the POV terms 'correct' and 'incorrect' in reference to this expression."

Things have come to a pretty pass when correct and incorrect are considered "POV". Is it POV to say 2+2=4 is correct and 2+3=4 is incorrect? There is not just one misinterpretation of the expression in question, and those misinterpretations that are incorrect are incorrect precisely because they clearly are misinterpretations. In particular, the notion that in this expression proves means tests is a complete fabrication.

Re: "The other point is that 'proves' is not equivalent to 'implies' or even 'infers'. "

Does it occur to you that implies is logically the opposite of infers? In any case, the expression in its original sense clearly means that the rule is to be inferred from the exception (or, conversely, that the exception implies the rule, which is another way of saying the same thing.)

Re: "This expression is far more subtle than it has been made out to be.":

It's really not very subtle at all. TheScotch (talk) 18:31, 6 July 2010 (UTC)Reply

In exceptio probat regulam in casibus non exceptis the meaning of probat can be taken to mean 'tests' rather than 'proves' in the modern sense - see etymology of 'proof' [[1]].

The result is that the meaning can be: "the exceptions test the rule". This was mentioned on a recent BBC Radio 4 programme (which I will reference as soon as I can locate it !) Mike0whit (talk) 19:53, 23 January 2012 (UTC)Reply

I agree that this is the sense - 'the exception tests (even 'probes') the rule' - i.e. it is the presence of exceptions which tests whether it is really a rule. This can be linked to Karl Popper's idea of a scientific theory as needing to be falsifiable. It may seem a general rule, but the exceptions will test it. Either it will be found that the exception disproves the rule, or that it is not a case which the rule prohibits, and therefore was misidentified. I guess someone needs to run of and find some sources, add to the article and attribute! Tsop (talk) 08:31, 1 November 2014 (UTC)Reply

The discussion of this in the opening paragraph of the article is utter garbage. It's not incorrect, it's just confusingly written garbage that belongs here, not in the body of the article. "The explanation that "proves" really means "tests" is, however, considered bogus by some sources.[2][3] To say "test " and "prove" is bogus, is immediately disproved by the existence of Engineering Test Facilities as Aberdeen Proving Ground and White Sands Proving Ground,both major proving grounds used for the test, proofing or proving of major USA military systems. The use is self-evident in the continuing use of these facilities to conduct engineering tests to prove or test the worthiness and readiness of engineering designs" — Preceding unsigned comment added by 137.132.217.9 (talk) 08:30, 28 June 2016 (UTC)Reply

The article begins by saying that use of this expression is "frequently confused." But from what I've heard, "frequently" means "almost always." Typically, one person claims something, then someone aks "but what about this counter-example?" and the response is that the counter-example is "an exception that proves the rule" which is claimed as support that their claim is correct, even though they are ignoring the counter-example, which they are not even trying to explain.

The article is good at explaining the logic behind this confusing phrase -- as with the "parking prohibited on Sunday" example (which would be strengthened by the "common law tradition" described above) -- but I think it would be improved by giving carefully designed examples of how this phrase should NOT be used, and why these uses are illogical and should be avoided. Craig Rusbult, craig@chem.wisc.edu (not registered as a wiki-contributor)

Fowler certainly believed that the fifth usage (which he called "Serious Nonsense" and regarded as incorrect) was the commonest, although he doesn't offer any evidence for this. I am not aware of any evidence that would support a choice between "frequently" and "almost always", and it would be pleasing to imagine that Fowler and others might have persuaded many people to avoid this usage, so that its frequency might be falling. Meanwhile, an earlier discussant (see above) took the view that labelling something as "incorrect", especially when so commonly used, expressed a Point of View (POV) and was therefore inappropriate for Wikipedia. So while I am sure most (if not all) style guides would agree with Fowler, and I personally regard Fowler as an invaluable guide to correct English usage, I am content with an article that merely reports Fowler's view and does not try to assert a would-be-universal view. --RichardVeryard (talk) 20:03, 7 October 2008 (UTC)Reply

I am concerned that the article has basically been taken over by Fowler's view. Brangifer (talk) 05:16, 3 August 2009 (UTC)Reply

An editor has complained that "this article or section relies largely or entirely upon a single source" - namely Fowler. I don't see a problem with this, given that many later style guides acknowledge Fowler as an influence, and I should be sorry to see this article made more cumbersome or less readable (or worse still, less informative) by the addition of further sources that merely copy or simplify Fowler's analysis. Repetition is not verification. --RichardVeryard (talk) 11:48, 4 December 2008 (UTC)Reply

I do see a problem with it. It gives undue weight to Fowler's view. This isn't an article about Fowler. Brangifer (talk) 05:16, 3 August 2009 (UTC)Reply

Of course it isn't an article about Fowler. But since every other style guide acknowledges Fowler and/or copies/simplifies Fowler's account, Wikipedia is merely reflecting the weight that countless other authorities have already accorded to Fowler. Unless you can find any source that is not clearly derived from Fowler. --RichardVeryard (talk) 10:32, 4 August 2009 (UTC)Reply

This is an odd end to the day, and a little off-topic really, but a good example. I was looking at a user page today, I forget why, and I noticed that she had a tag that said she objected to all violence against women.

So, exceptio probat regulam, violence against men is OK then?

SimonTrew (talk) 03:21, 26 April 2009 (UTC)Reply

The presence of a rule objecting to all violence against women does not imply the presence of a rule approving of all violence against men. Exceptio probat regulam would be satisfied in this case by the presence of a weaker rule disapproving of most violence against men. In general, an exception may indicate that some rule exists, but you cannot always infer the exact content of the rule from the exception alone. --RichardVeryard (talk) 21:31, 26 April 2009 (UTC)Reply

Actually, I think it is fair to assume that most persons who go on about "violence against women" are, in fact, relatively unconcerned about violence against men, however politically incorrect it may be to point this out. The possible presumption is that men are big and strong and able to defend themselves, although obviously they wouldn't be able to defend themselves very well against a gun, a gang, or a sneak attack. In any case, "the exception that proves the rule" is only a rule of thumb; it has plenty of exceptions itself that don't prove its rule. That doesn't give us license to invent some nonsense about proves in this case meaning tests. TheScotch (talk) 15:34, 7 July 2010 (UTC)Reply

Sorry to be both picky and slow in response, but I think TheScotch has missed the point of my comment. I agree that such persons may well be relatively unconcerned about violence against men. So we may infer the existence of some rule, but we cannot know precisely what rule. Thus even if the exception were to prove the existence of SOME (indeterminate) rule, it couldn't prove any specific rule. RichardVeryard (talk) 20:42, 19 January 2012 (UTC)Reply

There is a broader lesson to be learned from this. If the "parking prohibited on Sundays" example in the introduction is good colloquial usage, then the phrase "the exception proves the rule" applies to situations where nothing at all is proved. Just as "violence against women is bad" itself entails nothing about violence against men, so "parking on Sundays is bad" entails nothing about parking on any other day. We would normally accept the inference here that parking is fine except on Sundays simply because it is a common convention that parking is ok unless explicitly prohibited. However, there is no comparable convention about violence, so the inference to "violence against men is not bad" is unwarranted.

The broader lesson is: the expression "the exception proves the rule" applies only in situations where it is already clear that the presented exception is really an exception to some operative rule. Since this means we need to know beforehand that there really is a rule to which the postulated exception really is an exception, the exception itself plays no role in demonstrating the existence of the rule. All we get is:

1) P is an exception to some rule R;

Therefore,

2) Some rule R exists.

But this is a circular argument: the truth of (1) can only be established if (2) has already been established. The word "proves" seems to promise a demonstration of something less trivial than this, but the idiom is in fact totally vacuous and I avoid using it like the plague.80.42.223.155 (talk) 09:41, 28 November 2013 (UTC)Reply

Re: "In this second of Fowler's meanings, it is claimed the word 'proves' means 'tests' and that 'the exception proves the rule' therefore means something like 'an unusual case can be used to test whether or not a rule is valid'. If the rule stands up to the unusual case then that reinforces its truth; if not then the rule is disproved. However, this interpretation is not universally accepted: adherents of the original literal meaning maintain that an 'exception' here is not an extreme, unusual case, but merely any case that is not covered by the rule, and that 'proves' means 'demonstrates the existence of', not 'tests the validity of'."

Fowler mentions the theory that proves in the expression means tests not at all. If this article wants to discuss the theory that it means this in Fowler or anywhere else, for that matter, it has specifically to attribute this theory to an actual person, and not just through a footnote. Moreover, this actual person needs to be notable, and it needs to be notable that he says this.TheScotch (talk) 18:16, 6 July 2010 (UTC)Reply

I found a verifiable citation by chance in the Times some time last month. It is in the leader on page two, rather stupidly I tore off withour realising that the date was not on the top of the page but it was on a Saturday has a poppy on it so must be before Nov 14th one or two weeks at most, the edition number is 70098 so I am sure we can work back from there to find the exact saturday, I just only realised that I hadn't got the date on it, which is rather odd cos I tend to date everything anyway, I just suppose I assumed that it would be on the top of the heading and now find it is not.

Anyway the leader says, in second column, "The phrase 'the exception that proves the rule' [sic] originally meant that diffiult cases tested the probity of a generally held axiom."

The title for the leader is Court in the Act, discussing MP Phil Woolas.

Just adding for the record, it was very foolish of me not to realise the date wasn't on the top of the column, but I think that is certainly verifiable and may add to the article?

1) The Times is clearly wrong here.

2) The Times is not highly regarded for its writing, and Fowler's Modern English Usage is filled with example errors from this newspaper.

3) The passage quoted was stuck into the article obtrusively and illogically. TheScotch (talk) 08:07, 13 March 2011 (UTC)Reply

"frequently confused idiom" is really bad verbage. An idiom can't in and of itself be confused; it doesn't have a brain. skip that part and go directly to its meaning which also introduces the concept of confusion. Or cut frequently confused and go directly to the meaning. 4.249.48.173 (talk) 13:19, 12 January 2012 (UTC)Reply

The sentence isn't saying that the idiom is in a state of confusion, but rather that people confuse it frequently and thus, that it is confused (by people) frequently. That's how those kinds of constructions work. Animacy distinguishes one meaning from the other. It's fairly unambiguous to anyone who isn't trying to be an obtuse pedant. 108.70.133.22 (talk) 18:27, 13 February 2012 (UTC)Reply

Is this phrase really an idiom? The Wiki entry for idiom speaks of 'figurative meaning' : are not the words in this phrase used literally? GeorgeMogMorris (talk) 01:40, 24 December 2012 (UTC)Reply

Fowler's definition of Idiom is not the same as Wikipedia's, and allows for non-figurative idioms. However, Fowler doesn't describe exception-proves-the-rule as an idiom either, so I can see no good reasn to keep this word. RichardVeryard (talk) 08:48, 30 December 2012 (UTC)Reply

Thank you for this. I don't now know what to suggest on this matter and so will shut up for a time! GeorgeMogMorris (talk) 00:37, 16 January 2013 (UTC)Reply

Is the sense identified in the article as the 'loose rhetorical sense' actually the reinforcement of the idea that there exists a general rule (even if an imperfect rule; one having exceptions) by the realisation of how very unusual a case has to be for it to be an exception? If so, is there some better way of describing and naming this? GeorgeMogMorris (talk) 02:03, 24 December 2012 (UTC)Reply

The term "Loose rhetorical sense" comes from Fowler. I assume he wished to distinguish the first four senses in terms of different but legitimate purposes - legal, scientific, rhetoric and humour. I have no objection to anyone finding a better way of explaining this, but I suggest we also retain Fowler's term, perhaps in the form "Fowler calls this ...". RichardVeryard (talk) 09:11, 30 December 2012 (UTC)Reply

Thank you for this. I now understand more fully. I agree that it is desirable to retain Fowler's term and think your suggested form a good one. GeorgeMogMorris (talk) 00:37, 16 January 2013 (UTC)Reply

I would suggest that a less-political example be used instead of the Ted Kennedy example. The conclusion that 'the statement about rich, white politicians is not a rule, it is merely a trend' detracts from an article that I otherwise found very useful and informative. 76.241.123.147 (talk) 02:13, 13 February 2013 (UTC)Reply

From considerations from before reading the article I have a conception that might or migth not fall within the "scientific sense". That is if before the correction you have 9/10 rule holdings to rule applications and you have 9/9 rule holdings to rule applications you have in effect "boosted" the rule strength from 0.9 to 1.0. However if you have 9/10 (same as before) at start and then end up with 87/100 the rule "strenght" has gone down but its universality has gone up. The higher amount of thing to have been applied to means it more of a general patterns rather than anecdote. In the same sense that a 100% approval rating is likely to be a forged election result a finding that is only mostly universal is much more reliable than a exactly universal one if it of inductive "empirical" matter instead of definitional "deductive" matters (2+2=4 is sitll way more true than 2+2+x=5 even if x represents a value that might occasionally differ from 0).

Is this a thing? Should I search whether it is a thing to avoid it be original research? Do those different subsesenses count as things important enough to explicicate as different? 84.249.106.127 (talk) 09:10, 17 August 2015 (UTC)Reply

Hello fellow Wikipedians,

I have just modified one external link on Exception that proves the rule. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20080405180021/http://alt-usage-english.org/exception_proves.html to http://alt-usage-english.org/exception_proves.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 03:04, 26 September 2017 (UTC)Reply

Paleontologist and pop science writer Stephen Jay Gould defined this expression using the 'tests' meaning of 'proves.' As mentioned in this article, in this sense the expression means that when an exception to a rule is found, that exception can be used to test whether the rule really holds, or whether it needs to be modified or rejected. I believe that his example was the belief of Exploration Age Europeans that 'all swans are white.' In Australia, they found that there were black swans. So the rule that swans must be white (or at least the definition of swans as white birds) was in fact incorrect.

Gould pointed to the mathematical 'proof.' A proof in mathematics is a step by step demonstration that something is true. So a 'proof' is a test of a claim. I suspect that many people got this definition of the expression from Gould like I did, though he's not mentioned here.

It seems to me that both the legal principle and the Gouldian logical version are useful. Both could be used, with the context defining the particular meaning of the expression in that case.

Bryan Garner comes down on the side of the Latin/legal usage in his fine book Gardner's Modern English Usage, but Gardner isn't perfect. He calls 'self-confessed' reduntant, when it certainly isn't.

MarkinBoston (talk) 19:14, 13 February 2019 (UTC)Reply

The modern German verb "pruefen" indeed means "test". It would be nice if that were the end of the matter. But unfortunately the Anglo-Saxon version "profian" doesn't mean that, unless the lexicographers have over-hastily assumed for it the meaning of the Latin "probare". But if the matter in question is just Ine's laws [2], then perhaps we can admit ambiguity in their interpretation. FangoFuficius (talk) 12:13, 2 December 2021 (UTC)Reply

This page is for improving the article, not for chatting about the subject of the article. So, does this have any relevance for the article? If not, please go elsewhere. --Hob Gadling (talk) 13:41, 2 December 2021 (UTC)Reply

Atlas: All crows are black.

Zeus: Look, there’s a white crow!

Atlas: Well, that’s the exception that proves the rule.

Zeus: How does the existence of a white crow prove that all crows are black? It actually disproves it.

This in a nutshell is the problem with taking “prove” to mean mean confirm or establish. “Prove” means test, and in this case, the white or albino crow tests the rule that all crows are black, and demonstrates that the rule is actually: “All crows are black except for albinos.” 2600:4040:5D38:1600:50D8:B705:D849:1062 (talk) 15:52, 1 October 2022 (UTC)Reply

I think the theory is that an apparent exception tests the validity of the rule, and the rule either stands, falls, or is modified as a result. As the article indicates, there is not much evidence for this meaning, and it doesn't match how the phrase is used.--Trystan (talk) 17:13, 1 October 2022 (UTC)Reply

The opening refers to five ways, just as the source does. The description opening Uses in English describes the spectrum from the original to the most objectionable, and that most objectionable use, from the source, would be the going beyond the jocular nonsense described in 1.4 to the serious nonsense which should be 1.5. However, there is no 1.5 and the reader goes straight from an opening mentioning five ways to an article body that drifts off after just four ways. — Preceding unsigned comment added by 73.152.26.22 (talk) 19:32, 18 December 2022 (UTC)Reply

I thinks it's important in the introduction to present the differing explanations of this phrase as neutrally as possible. Some people think that there are "correct" interpretations of words and phrases, others think in terms of etymology and the roots of words and phrase. However, I would suggest that is better to be empirical and simply note that different usages exist.

I myself am etymologically minded and do use this phrase with the meaning that "prove" originally meant "test". I am not, though, saying that this is "correct". It is simply what appeals to me logically.

It is better to avoid making any judgments in the introduction which is why I have reworded the ending as I have and removed the reference to a paper which is not helpful at that point. By all means, it can be referenced later.

The paper concerned gives an opinion but is not an authority. There are no authorities on language usage. That people do use the expression believing "prove" means "test" is incontrovertible and does not require sources. The various discussions in the talk page are proof(!) Pooh bear138 (talk) 09:47, 15 June 2023 (UTC)Reply

I don't see any judgment made in the part of the lead you replaced, only statements of fact. So it is not clear to me what statement you're challenging. And you didn't just remove the reference to a paper, you removed all three cited. Nardog (talk) 10:20, 15 June 2023 (UTC)Reply

This article is not well written, overly verbose/convoluted and relies too heavily on a single source. It is in severe need of editing and simplification. I enjoy sandwiches (talk) 05:57, 26 November 2023 (UTC)Reply

Any suggestions? _{Duly signed,} ⛵ WaltClipper -(talk) 00:34, 2 December 2023 (UTC)Reply

I read it as being concise and well-explained. 91.181.209.21 (talk) 21:25, 9 May 2024 (UTC)Reply