What's the meaning of the expression, "That's the exception that proves the rule?"

As I understand, the expression, “The exception that proves the rule,” is based on Dr. John Snow’s investigation of cholera cases. He noted that all the cholera cases were in people who lived near an apparently infected pump, except for one, the exception. But on closer inspection the woman turned out not to be an exception after all as she took her water from the infected pump.
Thus, the more accurate phrase would be:
“Only apparent exceptions, that turn out not to be exceptions, prove the rule.”

I’ve never heard of the expression having anything to do with Dr Snow, and no, it doesn’t make any sense that it should.

Here’s Michael Quinion’s explanation of the phrase. See what you think – it makes sense to me as he explains it.

The column in question.

So to summarize, the explicit statement of an exception indicates that the rule applies in all other circumstances.

What this proverb implies is that there are no exceptions to a rule except those that are explicitly stated - nearly the opposite of the anti-logical meaning it is given today.

It would be nice to adapt the saying to describe the important principle of falsifiability, but I can’t see any neat way to do it myself. “The logical possibility of an exception proves the well-formedness of the rule” is the best I’ve got.

I think it is just used to counter people that are doing their best to find the one exception tp your general statement. Like when any sane person would agree that the German football team plays boring, defensive foorball; someone will come and say they played nicely in 2006. This was true, but doesn’t change the fact they played boringly for more than 50 years before…

(now I really hope some german football fans show up;))

The usual form of this old aphorism is “The exception proves the rule.” You’ll never make sense of this expression until you consider the earliest meaning of the word “prove.” To “prove” originally meant to test. So what this expression is saying is that an apparent exception will test the correctness of a rule. If the rule fails to hold when tested by an exception, then the rule is incorrect.

Did you read the column, or the link to Michael Quinion? Your definition doesn’t make sense for the original meaning derived from Roman law, and it doesn’t make sense for the current, nonsensical “any exception just makes the rule stronger”.

The way I’m reading that column, Cecil ends by conceding that the ‘tests the rule’ meaning is pretty much on the money.

…apart from the little difficulty that that is not what the column actually says, which is that the Civil-Law origin appears to be correct, and the “tests” theory is merely almost true.

For comparison, the Norwegian equivalent, apparently from the same latin source, litterally translates to “The exception that confirms the rule.”

I accept Quinion’s explanation, although I previously thought WordMaven’s was correct. But it remains true that once upon a time “prove” meant “test”. The 1962 translation of C.F. Gauss’s Disquisitiones Arithmeticae, translated by A.A. Clarke, SJ, has a bad mistranslation caused by the failure to take this change of meaning into account. In discussing a certain theorem (I think it was that every positive number is the sum of four (or fewer) squares), he has Gauss saying that Langrange proved this by induction, but we demonstrate it by infinite descent. Today, “prove by induction” and “demonstrate by infinite descent” are synonymous phrases. What Gauss was really saying was that Lagrange had tested the assertion in many cases and concluded (by scientific, not mathematical) induction that it was likely true. In fact, Gauss gave the first complete and correct argument.

The old sense of “prove/proof” still lives in many contexts, such as[ul]
and formerly (but I don’t know if it still is) the manufacture of explosives.

But, of course, the fact that the meaning exists does not prove (sorry!) that it is responsible for the idiom in question.

One further comment. The reference is said to be to Civil Law. Note that Civil Law is not used in England or most of the United States, but is used in Scotland and the State of Louisiana (and Québec, too, I think). That’s just the sort of thing to establish the maxim in English, while leaving it obscure in most of the English-speaking world.

John, can you clarify what you mean by “Civil Law”, and why it is not used in England and most of the United States? Because there are two (well, really three, depending upon the jurisdiction yada yada) types of legal cases: criminal and civil. Wouldn’t a civil court case be “Civil Law”?


No, “Civil Law” has nothing to do with “civil case”.

Throughout the Western world, and in a good many former Western colonies as well, the skeletons of the various legal systems ultimately descend from the late Roman Empire. (No kidding!) But, for a number of historic reasons, including Danish invasions and church-state relations in the middle ages. the law in England went off in several unusual directions. (Two important ones are the central importance of precedent and the splitting off of Equity into an independent and parallel system.) The English system came to be known (in English) as “Common Law”, and is distinguished from “Civil Law”, which means all Roman-based legal systems that are not Common Law. Common Law would have been little but a quaint eccentricity if it had not been for the power and scope of the British Empire, but, as history worked out, Common Law became the legal system of America, Canada, Australia, New Zealand, Ireland, India, and much more. Even nations that have disassociated themselves from England have generally retained Common Law, because, in the modern world, dropping it would be catastrophic; it would essentially involve rewriting every single law, and redrafting every contract, deed, will, and what have you – and only after every lawyer and every judge went back to school for two or three years.

There are a few Civil-Law enclaves in the Common-Law world. [ul]
[li]When Scotland and England were officially joined to form the United Kingdom in 1707, Scotland insisted on keeping Civil Law. For nearly three hundred years, the Parliament in London had to keep up two completely different sets of laws, one for Scotland, and one for everyplace else. (Scotland has its own Parliament nowadays.)[/li][li]After Britain conquered Canada in 1763, they allowed Québec to keep Civil Law, as part of the general program to keep the French Canadians happy, which also involved, for example, allowing Roman Catholics in Québec full civil rights, which Roman Catholics in England would not receive for generations.[/li][li]When the US purchased Louisiana from France in 1803, the civilized part of Louisiana, which eventually became the State of Louisiana, asked to retain Civil Law, to the extent that the US Constitution allowed it.[/li][/ul]

Because of Scotland, Québec, and Louisiana, Common-Law lawyers have to know a little about Civil Law (at the very least, they need to know some phone numbers).

With one huge exception that, while not proving any rule, was historically very important. English ecclesiastical courts are civil law jurisdictions. (The same was also true of a number of smaller courts, such as the court of admiralty.) This is why before the late eighteenth century, law degrees from Oxford and Cambridge were only in civil law. Civilian lawyers were a distinctive subset of the London legal profession and, until the nineteenth century, the church courts handled a wide range of seemingly secular cases. A familiarity with civil law was therefore rather more common in England than you might suppose.

Most likely because they had been Catholic canon law tribunals (before Henry VIII had his little snit fit), and Catholic canon law is just about totally based on Civil Law principles.



When rendered in the upper case “Civil Law” has nothing to do with “civil cases”, but civil cases are all about lower case “civil law”, i.e., “not-criminal law”, a distinction which the editors of Wikipedia get:

In common use the term is ambiguous, and you have to grok from context whether someone is using “civil law” to mean “system based on Roman law” or “not-criminal law”.




Thanks for the clarification.

Whew! I didn’t know that, though I can see it. (I should add that I’m not even an eensy bit a lawyer or a historian.)