The Volokh Conspiracy - "The Exception Proves the Rule":

People often find the origin of this phrase puzzling -- wouldn't an exception disprove the rule? Some have argued that "proves" was used in its meaning "To make trial of; to try, test" (definition II in the Oxford English Dictionary).

But as best I can tell, the origin of the phrase is the legal principle that the statement of an exception shows that the rule is opposite in the cases not excepted. Here's what the OED tells us:

The legal maxim, ‘Exception proves (or confirms) the rule in the cases not excepted’ (exceptio probat regulam in casibus non exceptis), which is in its original form an example of sense 1, is commonly quoted as ‘The exception proves the rule’ ....

Here's an early English American case, Watson v. Alexander, 1 Va. 340 (1794): "The act excepts the cases of contracts for gold or silver coin, tobacco, or other specific property; and if it be true, that an exception proves the rule, we must decide that all other contracts are within the law." And here's an illustration given by the OED that shows the meaning beyond the legal context; it's from Samuel Johnson's preface to a 1765 edition of Shakespeare (emphasis added):

It has been much disputed, whether Shakespeare owed his excellence to his own native force, or whether he had the common helps of scholastick education, the precepts of critical science, and the examples of ancient authours.

There has always prevailed a tradition, that Shakespeare wanted learning, that he had no regular education, nor much skill in the dead languages. Johnson, his friend, affirms, that he had small Latin, and no Greek; who, besides that he had no imaginable temptation to falsehood, wrote at a time when the character and acquisitions of Shakespeare were known to multitudes. His evidence ought therefore to decide the controversy, unless some testimony of equal force could be opposed.

Some have imagined, that they have discovered deep learning in many imitations of old writers; but the examples which I have known urged, were drawn from books translated in his time; or were such easy coincidences of thought, as will happen to all who consider the same subjects; or such remarks on life or axioms of morality as float in conversation, and are transmitted through the world in proverbial sentences....

There are a few passages which may pass for imitations, but so few, that the exception only confirms the rule; he obtained them from accidental quotations, or by oral communication, and as he used what he had, would have used more if he had obtained it.

The Comedy of Errors is confessedly taken from the Menæchmi of Plautus; from the only play of Plautus which was then in English. What can be more probable, than that he who copied that, would have copied more; but that those which were not translated were inaccessible?

By "the exception confirms the rule," Johnson seems to have meant that "seeing the exception, and recognizing that it is an exception, confirms for us that there is a rule." And my sense is that in the modern day, the phrase -- when used effectively -- has roughly that meaning.

(link)

Dennis Nolan (mail):

Check the etymology. "Prove" used to carry the sense of "test." We have remnants of that earlier meaning when we speak of "proving a theorem" or refer to a "proving ground." The phrase thus refers to an exception that tests the rule, meaning one that draws the boundaries of the rule by showing where it might not apply. Because we've nearly lost that meaning, almost everyone interprets the phrase as you do --- including judges who should know better.

5.8.2009 3:08pm

JB:

I've always understood it to mean "This is an example of X, so the rule should normally apply, but it has Z exceptional quality and thus the rule not applying is understandable."

I.E, Barack Obama is an exception to the rule that Americans won't vote a Black man for President, but he proves the rule because his life story, and the political context of the campaign, are so unusual and many of the events that had to go his way for him to win were dependent on those unusual aspects. So his election shows just what combination of bizarre circumstances are necessary for a Black man to be elected, and since those circumstances are unlikely to recur the election of another Black President remains very unlikely.

Actually, having read Dennis Nolan's comment, I think he states my meaning more concisely. The exception proves, i.e. defines the limits of, the rule, by showing what exceptional qualities are required to set it aside.

5.8.2009 3:10pm

Houston Lawyer:

I always it referred to the rule of "there is an exception to every rule".

5.8.2009 3:12pm

Michael A. Koenecke:

Folks, per the inestimable Cecil Adams, the phrase goes back to Cicero, and Professor Volokh's definition of its meaning is correct. See http://tinyurl.com/67zznr .

5.8.2009 3:21pm

boose:

I feel like this is a case of too many lawyers thinking like lawyers. I always thought "the exception proves the rule" comes from people expecting to find exceptions to rules. I don't think anyone takes the phrase seriously (none that I know). They're just joking that the existence of an exception proves that the rule is true.

5.8.2009 3:28pm

Splunge:

I think Professor Volt is quite correct. The existence of an exception, recognized as such proves the existence of a rule.

Or, to put it another way, if X surprises you, it follows that not X is the rule, or more generally, that there is a rule predicting the probability of X.

I don't think the saying is meant to distinguish between differing rules, but to distinguish between situations of chaos, where no rule is known, to situations of regularity where the rule has been forgotten.

Frequently enough, people forget or ignore basic rules, for all kinds of social and psychological reasons. For example, it is entirely probable that a white man in 1855 would have absorbed racism with his mother's milk, so to speak, and think that the inferiority of black men in certain areas was obvious, and certainly not worthy of introspection and asking why.

But if he were confronted with a successful and competent black President, should one be elected someday, he would be surprised. That surprise should, the folk warning goes, be sufficient proof to him that there is an unexamined rule operating in his life.

5.8.2009 3:29pm

Here's the comment in the link Michael posted. As it shows, the common legal understanding of that phrase is wrong because the Latin root word, probat, means prove in the sense of "test" as I explained above. The link doesn't say that the common usage is right, just that it's common despite being wrong.

Don't you get it? The whole point of this saying is that it doesn't make sense. It's what you say to confound your enemies when your argument has been shot out from under you by some pesky counterexample. From the point of view of advancing the debate it's about one jump ahead of "yo mama," but it beats standing there with your mouth open.

To be sure, a few scholarly types have tried to make excuses for "The exception proves the rule," as the quotation books usually phrase it. They say it comes from the medieval Latin aphorism Exceptio probat regulam. Probat means "prove" in the sense of "test," as in "proving ground" or "the proof is in the pudding." So "the exception proves the rule" means a close look at exceptions helps us determine a rule's validity.

If Latinists understand it that way, however, they're pretty much alone. I've looked up citations of this saying dating back to 1664, and in every case it was used in the brain-dead manner we're accustomed to today--that is, to suggest that non-conforming cases, by the mere fact of their existence, somehow confirm or support a generalization. Obviously they do nothing of the kind. We like to think proverbs become proverbial because they're true; this one is an exception. It certainly doesn't prove the rule.

5.8.2009 3:34pm

Sean O'Hara (mail) (www):

The example I always use when trying to explain this to people is a sign that says, "No parking from 5 AM-5PM Mon-Fri" -- the rules generally allow parking, except in the periods noted.

5.8.2009 3:36pm

Robert Franson (mail) (www):

Excellent! This is a phrase that has long seemed to me a niggling reversal of common sense, like tumbling head over heels.

Your discussion also now reminds me of Karl Popper's falsifiability:

Popper take[s] falsifiability as his criterion for demarcating science from non-science: if a theory is incompatible with possible empirical observations it is scientific; conversely, a theory which is compatible with all such observations, either because, as in the case of Marxism, it has been modified solely to accommodate such observations, or because, as in the case of psychoanalytic theories, it is consistent with all possible observations, is unscientific.
Karl Popper: Stanford Encyclopedia of Philosophy

So the simple property of being able to discern an exception suggests that we have a thought-out rule rather than an emotional edict.

5.8.2009 3:37pm

geokstr (mail):

JB:
So his election shows just what combination of bizarre circumstances are necessary for a Black man to be elected, and since those circumstances are unlikely to recur the election of another Black President remains very unlikely.

Well, I don't think it will be possible to test your exception until at least 2036, the first year that Malia will be eligible to run against him in the primaries (assuming she will even be able to disprove his legal challenge that she is not a natural born citizen.)

5.8.2009 3:38pm

greenish (mail):

Sounds like the same principle which resulted in the 9th Amendment. They worried (and history proves them right to have worried) that the Bill of Rights might have looked like the exception to the rule "the Federal Government can make any laws".

5.8.2009 3:41pm

Jonathan F.:

And my sense is that in the modern day, the phrase -- when used effectively -- has roughly that meaning.

I see you've found a sufficiently opaque, though nearly equivalent, way of saying "when used correctly." Well done! I wonder if commenters will remember this phrase the next time you condemn prescriptivism -- after all, who could be against effectiveness?

5.8.2009 3:46pm

Fub:

Houston Lawyer wrote at 5.8.2009 3:12pm:

There is an exception to every rule.

Therefore there is an exception to that rule.

Therefore there is some rule to which there is no exception.

Therefore there is not an exception to every rule.

I think I need a nap.

5.8.2009 4:05pm

rick.felt:

The Wikipedia entry lists five interpretations, in order of plausibility. The "proves = tests" interpretation is disfavored, and Prof. V's is favored.

5.8.2009 4:08pm

CJColucci:

The example I always use when trying to explain this to people is a sign that says, "No parking from 5 AM-5PM Mon-Fri" -- the rules generally allow parking, except in the periods noted.

Similarly, when I'm eating at a restaurant and see the familiar sign saying "Not responsible for personal property," I feel reassured because the restaurant is thereby taking responsibility for real property and I don't have to worry about my house burning down while I'm out.

5.8.2009 4:23pm

Bama 1L:

I'd learned it was from Cicero's Pro Balbo:

Quod si exceptio facit non liceat, ubi non sit exceptum, ibi necesse est licere.

My translation:

If an exception makes it impermissible, where there is no exception, it must be permissible.

So if there is a stated exception, then there must be a general rule to the contrary--the exception proves the rule.

But I don't think this describes all instances of actual usage.

5.8.2009 4:27pm

CJColucci, I think Cicero would say that the restaurant's disclaimer of responsibility for personal property means that there is a general rule that restaurants are responsible for personal property.

But that is less funny so yours is better.

5.8.2009 4:29pm

Bob White (mail):

The invaluable Michael Quinion, to whose World Wide Words newsletter you really should be subscribing (yes, it's free), has addressed the issue.

My experience suggests that most people using the rule to refer to an exception do it the way I used to, in that the specific example is an exception to a known rule of general application. In this sense, it only backhandedly conforms to the correct meaning of the phrase.

5.8.2009 4:31pm

Julian Sanchez (mail):

I had always interpreted it as Eugene does here, and that seems to me to make perfect sense, even outside the legal context. If I note that poetry and literature were largely the province of men in the ancient world, and you immediately come back with "but Sappho!" — that's an "exception that proves the rule": You remembered Sappho immediately just because an ancient Greek female poet was rare and memorable. And there are plenty of cases where most people can think of one or two counterexamples to a general claim, but the very reason the counterexamples are both widely known and rapidly identifiable AS counterexamples is that they're unusual.

5.8.2009 4:42pm

ras (mail):

I too agree w/Eugene's interpretation, and would add that this saying goes, to me, hand in hand with "rules were made to be broken." Both indicate a willingness to use judgment in the application of the rule, a.o.t., say, a zero-tolerance attitude. The former humbly admits that we cannot anticipate everything; the latter insists that we can.

5.8.2009 4:49pm

Frater Plotter:

The meaning is very simple. Let's say question disputed is whether you are allowed to park next to the post office on Monday nights, when the posted sign says: "No parking here on Wednesday mornings".

The sign doesn't say anything about Monday nights, but there would be no point saying "No parking here on Wednesday mornings" unless the default case (the "rule") was that parking is allowed.

When the law spells out an exception (here, the time when you may not park next to the post office), it thereby implies the existence of a more general rule to the contrary (here, that at any other time, you may park there).

Why is this important? Because it improves public knowledge of what is permitted. You don't have to define by extension all of the times when it's okay to park; you just have to define the exceptions, which define the rule as a negative space.

We can derive this idea also from Grice's maxims of implicature. If parking were never allowed, then putting up a sign saying "No parking on Wednesday mornings" would be perverse and misleading.

5.8.2009 4:55pm

David Schraub (mail) (www):

One of the greatest entries in the all-around fabulous Devil's Dictionary by Ambrose Bierce, for "exception", dealt with this precise topic:

"The exception proves the rule" is an expression constantly upon the lips of the ignorant, who parrot it from one another with never a thought of its absurdity. In the Latin, "Exceptio probat regulam" means that the exception tests the rule, puts it to the proof, not confirms it. The malefactor who drew the meaning from this excellent dictum and substituted a contrary one of his own exerted an evil power which appears to be immortal.

5.8.2009 5:03pm

Crust (mail):

What Jonathan F said. I detect a hint of prescriptivism.

5.8.2009 5:13pm

deenk:

Logically, an exception disproves a general rule, unless its exceptional status is misleading. The example of Shakespeare using the English translation of Plautus as a model (rather than an untranslated text) is a good illustration.

Another one is that workers in tiger-infested parks in India wear human masks on the backs of their heads, following the rule that tigers invariably attack humans from behind. Nevertheless, a worker was killed by a tiger. It was found that he had removed his mask in order to eat lunch (but, sadly, he became lunch). This is an exception that proves (adds further support to) the rule.

I admit that the phrase, as commonly used, is twaddle.

5.8.2009 5:21pm

Eugene Volokh (www):

Jonathan F, Crust: I meant to say that when it's used in a way that the listener nods and says, "yes, that's right"; Julian Sanchez offers a good example.

Now it might be the case that "exception proves the rule" is mostly used as just a substanceless dodge (deenk's "twaddle") to explain away an exception, for instance "The Fourth Amendment requires probable cause and a warrant for all searches." "But how about automobile searches, which don't require a warrant." "The exception proves the rule." That hasn't been my experience, but I might be wrong. So I wanted to qualify my assertion about usage, to focus on those uses that do make the listener perceive that there's a substantive argument behind the figure of speech, rather than just a rhetorical dodge.

5.8.2009 5:35pm

non-native speaker:

The exception confirms the rule, as we say in my country.

5.8.2009 5:57pm

cboldt (mail):

Apologies to all who expressed this before I did: If there isn't an exception, there isn't a rule.

5.8.2009 5:59pm

I need to rephrase. If the situation isn't being argued as an exception, then the argument isn't about a rule.
.
Or, without the presence of a rule, there is no such thing as an exception to the rule.

5.8.2009 6:05pm

Mike G in Corvallis (mail):

It's a useful maxim, either way, but does Eugene's interpretation of "the exception proves the rule" have any force in American law? I seem to recall a discussion here a few months ago about the Fifth Amendment:

No person shall be held to answer for a capital, or otherwise infamous crime, unless on presentment or indictment of a Grand Jury, except in cases arising in the land or naval forces, or in the Militia, when in actual service in time of War or public danger; nor shall any person be subject for the same offense to be twice put in jeopardy of life or limb; nor shall be compelled in any criminal case to be a witness against himself, nor be deprived of life, liberty, or property, without due process of law; nor shall private property be taken for public use, without just compensation.

If Eugene's interpretation holds, then doesn't the text in boldface imply that after due process of law, the government can take private property for non-public use without any compensation to the owner?

(Alas, if my vague recollection is correct and this issue was discussed on the Conspiracy, I don't recall the resolution — if any.)
)

Don't you get it? The whole point of this saying is that it doesn't make sense. It's what you say to confound your enemies when your argument has been shot out from under you by some pesky counterexample.

So the Romans invented the Chewbacca Defense?!
›

5.8.2009 6:11pm

Just thinking out loud, perhaps, but if the original English translation had been "the exception proofs the rule" - similar to proof as in proofread, test, etc - then the current phrasing that instead uses proves might have evolved as a mere phonetic twist. Many other expressions have done this, such as when "dressed to thine eyes" morphed into "dressed to the nines", and this could be the same process at work.

5.8.2009 6:23pm

Many other expressions have done this, such as when "dressed to thine eyes" morphed into "dressed to the nines", and this could be the same process at work.

That seems also to have been the case with "Brightness falls from the air".
›

5.8.2009 6:47pm

DennisN (mail):

@ Fub:

Houston Lawyer wrote at 5.8.2009 3:12pm:

I always it referred to the rule of "there is an exception to every rule".

There is an exception to every rule.

Therefore there is an exception to that rule.

Therefore there is some rule to which there is no exception.

Therefore there is not an exception to every rule.

I think I need a nap.

I suspect that even Cicero understood it in that sense, even if the statement had a more formal purpose.

I've never taken it as a completely serious statement. It's more of a wiseass comment that "The rule you're citing doesn't really apply in this case, but I can't be bothered to develop an argument proving that."

5.8.2009 7:13pm

seadrive:

JB: I've always understood it to mean "This is an example of X, so the rule should normally apply, but it has Z exceptional quality and thus the rule not applying is understandable."

This can work two ways. If we state the proposition as "A is the set of all X". Then we can have an X which is not in A, or some element of A which not an X.

To be concrete, we can say that for some track, all race cars can lap in under one minute, and only race cars can lap in under one minute. One exception would be a slow race car; the other would be a very fast non-race car. Either would be an exception, but analysis of the exception would reinforce acceptance of the general rule.

5.8.2009 7:42pm

Dennis Nolan: The Oxford English Dictionary people can certainly err, but I tend to be inclined to accept their judgment unless there's some pretty concrete authority to the contrary. Can you point, please, to some such specific authority that demonstrates that their tracing of the etymology, and their translation of the Latin, is mistaken?

5.8.2009 8:35pm

David Chesler (mail) (www):

I prefer "This apparent exception will force us to test if the rule is really true."
I accept Prof. Volokh's interpretation, and one of the last times this came around we talked about the ship's log entry "The First Mate was sober today."
But the most common usage I hear is along the lines of "The existence of a counter-example demonstrates that the purported rule is true." For instance, I state as a rule "All birds fly south for the winter." You say "That's not true, there are a bunch of Canada geese that won't ever leave, and I see seagulls all winter, and crows, and pigeons." I say "Well then it must be true, because the exception proves the rule." (Or as Julian Sanchez says, the only reason you can think of those birds is because they are so exceptional.) And it's just not so. Some birds migrate, and some don't.

As for the parking signs, it doesn't work that way in real life, at least to the degree that New York City represents real life. For instance my car was towed, on a Saturday afternoon, from a spot on Eighth Avenue and 35th Street marked “No Parking 2am - 6am Monday Wednesday Friday”. (That was bidirectional; the other direction of the sign from where I parked was NYP only.) At the impound lot I met someone whose car had been towed from the same place. Turns out at the other end of the block, under construction scaffolding, were signs that said "No parking 2am - 6am Monday Wednesday Friday" and "No Standing except Commercial Vehicles. Metered Parking 3hr limit 7am-7pm except Sunday". (I don't know what that means - do trucks have to put money in the meter to stand? Can trucks not only stand but also park for three hours?) (I'm out $200, but it was all for good, details on request.)

5.8.2009 9:55pm

In Soviet Russia, exception proves you.

(Eugene, can you confirm?)

5.8.2009 10:02pm

vidkunquisling:

Bierce's cynicism was always justified, and one would conclude that the error is indeed immortal from the numerous rationalizing posts here for a saying that made sense in Latin but has become nonsensical as the English cognate's meaning shifted over time. "Probare" means to test in Latin. Its original sense survives in "prove's" etymological cousins, "probe" and "probation."

5.8.2009 10:54pm

eyesay:

JB wrote

Barack Obama is an exception to the rule that Americans won't vote a Black man for President, but he proves the rule because his life story, and the political context of the campaign, are so unusual and many of the events that had to go his way for him to win were dependent on those unusual aspects. So his election shows just what combination of bizarre circumstances are necessary for a Black man to be elected, and since those circumstances are unlikely to recur the election of another Black President remains very unlikely.

That was, in the words of Spock, “a torrential flood of illogic.”

1. JB offers no evidence that a significant fraction of the American voting population would never vote for a black man (or woman), and more important, to the extent that some voters were not ready to vote for a black candidate in 2008, the fact that their worst fears of Obama, e.g. establishing some kind of black-preferential state, will not come true will help this racist remnant overcome their unwillingness to vote for a black candidate.

2. Barack Obama's method of winning the Democratic nomination and the election was not particularly bizarre. He communicated, organized, and fundraised effectively, and his methods are likely to be copied by others on both sides of the aisle.

3. Until 1960, America had never elected a Catholic president, but there's no reason to think that America wouldn't elect a Catholic president again, and there is no reason to believe that Sen. John Kerry's Catholicism cost him the election in 2004. JFK knocked down the Catholic barrier, once and for all. Geraldine Ferraro knocked down the barrier of a woman on a major-party ticket, once and for all. Barack Obama knocked down the barrier of a black major-party candidate and a black president, once and for all.

5.9.2009 2:30am

traveler496:

I was mulling this over last week (usually the latency is a day or two; not sure why it took so long this time..:-)

One can ask about this phrase's origin, its commonly intended meanings, or its useful meanings. Thanks to this thread I now know three useful meanings, exemplified respectively by:
- the OED &Cicero examples
- the Samuel Johnson example (though Johnson used 'confirms')
- the Sappho (noteworthiness-of-counterexamples) example

The interpretations wherein "proves" means "tests" or "defines the boundaries of," while coherent, don't strike me as very useful.

Attempts to equate or otherwise relate this saying to the (false) saying "every rule has an exception" seem even less useful, and muddled besides.

5.9.2009 2:37am

Dr. Weevil (mail) (www):

Probare does mean 'test' in Latin, but it also means 'approve' and 'prove'. For the last, see meaning 7 in the Oxford Latin Dictionary (p. 1465), with at least five examples from Cicero.

5.9.2009 1:50pm

Bob in SeaTac (mail):

I always understood it as determing whether the statement is a rule or a law.

If there is an exception, it is just a rule, not an inviolable law, such as the law of gravitation, to which there are no exceptions.

The spelling rule, 'i' before 'e,' except after 'c,' or when sounded as 'a' in 'neighbor' and 'weigh' is only a rule. There are many exceptions, one of which is 'science.'

If it were a spelling law, there could be no exceptions.

5.9.2009 7:47pm

Sniped by Bob in SeaTac ;-)

Your distinction between "rule" and "law," is worthwhile.

In either case, if there is an exception, it shows that the rule or law is insufficiently precise or the underlying science insufficiently understood, for the rule to be applied in this particular case. The fault is not with the exception, but with the rule.

In science, understanding those exceptions can help us present better theories. They can have a similar function in Law.

5.10.2009 2:06pm

KevinM:

Bryan Garner, who, since the abdication of Clapton, is god, says that "proves" in the sense of "tests" the rule is a spurious bit of folk etymology that doesn't reflect the actual background of this expression.
According to G, it originated as one of those Latin legal maxims: "exceptio probat regulam in casibus no exceptis" (the exception proves, or confirms, the rule in the cases not excepted). "Exception," in context means "the act of exception," not "that which is excepted." Thus, in a sense, its a rule of legal or even statutory interpretation. If an exception must be made, then it is implied that a more general (and contrary) rule must exist. If the swimming pool goes out of its way to post a sign that bathing caps need not be worn on Tuesdays, we may infer that there's a general rule that they must be worn.

5.11.2009 1:52pm

NON exceptis, not "no" exceptis - sorry.

5.11.2009 1:53pm

If you have a comment about spelling, typos, or format errors, please e-mail the poster directly rather than posting a comment.

Comment Policy: We reserve the right to edit or delete comments, and in extreme cases to ban commenters, at our discretion. Comments must be relevant and civil (and, especially, free of name-calling). We think of comment threads like dinner parties at our homes. If you make the party unpleasant for us or for others, we'd rather you went elsewhere. We're happy to see a wide range of viewpoints, but we want all of them to be expressed as politely as possible.

We realize that such a comment policy can never be evenly enforced, because we can't possibly monitor every comment equally well. Hundreds of comments are posted every day here, and we don't read them all. Those we read, we read with different degrees of attention, and in different moods. We try to be fair, but we make no promises.

And remember, it's a big Internet. If you think we were mistaken in removing your post (or, in extreme cases, in removing you) -- or if you prefer a more free-for-all approach -- there are surely plenty of ways you can still get your views out.