Behind the Dictionary
Lexicographers Talk About Language
True Facts and False Facts
When I was in elementary school, and was being taught the difference between fact and opinion, I wondered how to classify statements like "There are 51 states in the United States." It wasn't true, so it wasn't a fact, but on the other hand, it didn't seem to involve the kind of judgment that we were learning to identify with opinions.
I don't remember what the teacher said about it, or if the issue even came up in the classwork or quizzes. It might have just been a question in my own head that I never ventured to ask aloud. Evidently, though, this issue is now addressed in some language arts textbooks. In an episode of their radio show "A Way with Words" back in September, Grant Barrett and Martha Barnette spoke with a listener from Dallas, whose eight-year-old daughter had been taught that a fact was something "that could be proven true or false." Just recently, my son Doug reported the same thing for an online English class he was taking. My statement about 51 states in the United States would be called a "false fact."
But isn't false fact a contradiction in terms? Or to look at it from a different angle, isn't true fact a redundancy? That was certainly the view taken by Grant and Martha's caller, and indeed, Grant and Martha, too. Even Jonathan Lighter, the author of the Random House Historical Dictionary of American Slang, posted a message to the American Dialect Society's email list in 2005, taking issue with this more inclusive definition of fact.
As it turns out, though, the definition of fact as "a statement that can be proven true or false" has been active for four centuries. On the other hand, it has been criticized and contested for at least the last two centuries.
The earliest attestation for fact in the Oxford English Dictionary is from 1545. The word is a borrowing from Latin, deriving from the past participle of the verb facere meaning "make" or "do." The spelling shows that it came straight from Latin, not via Old French, and its original meaning was the one from Latin: "something made or done." The meaning of "something that is known to be true" came during the following century. The closest OED definition to "a statement that can be proven true or false" is its fifth one: "Something that is alleged to be, or conceivably might be, a ‘fact'." This one dates from around 1730. The OED even has an attestation of false fact from 1824.
Google Books allows a much earlier attestation of false fact, from 1688 in the Journals of the House of Commons. A passage reads:
Or, suppose the Judges were corrupt; and directed false Law, or false Fact, or overawed the Jury; or admitted a Party to be a Witness, or Juror (as in Effect it was); That was enough to render the Verdict corrupt….
Here's another one, from 1738, with a bonus true facts in the same sentence:
They will be of opinion that it is the duty of every corrector of a book to take from it all the false facts and to substitute true facts in their room ….
The earliest complaint I've found about using fact to refer to something that could be either true or false is from the Irish politician and reformer Daniel O'Connell. In a sensational trial regarding a conspiracy to assassinate a magistrate and several landlords in County Cork, O'Connell demolished the testimony of the prosecution's witnesses, and continually attacked the Solicitor General making the case. This trial is written about in a number of biographies of O'Connell, and one incident features in most or all descriptions of it: At one point, the Solicitor General said, "The allegation is made upon false facts." O'Connell pounced on this statement. Quoting from one of the biographies:
"False facts!" cried O'Connell. "What a bull! How can facts be false?" The Solicitor General retorted, "I have known false facts and false men, too."
An even more scornful rejection of the idea of false facts is found in the February 1839 issue of The United States Magazine and Democratic Review. The writer of an article on the banking system responds to a criticism of his earlier article, in which the reviewer deprecates the writer's collection of "true and false facts":
As to our collection of true and false facts — our want of knowledge of this deep and important distinction must be attributed to want of instruction as to the mode by which the Philadelphia school distinguish a true fact from a false fact. It may be trivial or important, but a fact is neither more nor less than a fact with us. A great favor would be conferred on the uninitiated, if the professors of that brilliant school would explain, when they assert facts, when false facts are intended. This would doubtless have saved much of the time employed upon this Article.
Aside from the fact (the true fact, in fact) that fact has been able to denote something either false or true for most of the time it has been in the English language, there is another possible argument to be made in favor of this more liberal definition. In its FAQ, the Usenet newsgroup alt.usage.english has entries for many favorite usage questions or complaints, and defends the expression true fact, first by pointing out that what might be redundant in one context might not be redundant in another; for example:
"Consensus of opinion" is redundant in most contexts, but not in "Some of the committee members were coerced into voting in favour of the motion, so although the motion represents a consensus of votes, it does not represent a consensus of opinion."
Fair enough. "True facts, not false myths" is defensible on these grounds. But what about the supposed contradiction of false fact? The entry continues:
Context can negate part of the definition of a word. "Artificial light" is light that is artificial (= "man-made"), but "artificial flowers" are not flowers (i.e., genuine spermatophyte reproductive orders) that are artificial. In the latter phrase, "artificial" negates part of the definition of "flower". The bats known as "false vampires" do not feed on blood: "false" negates part of the definition of "vampire".
The ordinary definition of "fact" includes the idea of "true" (e.g., fact vs fiction); the meaning of "fact" does have other aspects (e.g., fact vs opinion). Context can negate the idea of "true". …
It follows that "true fact" need not be a redundancy.
Interesting reasoning, but it raises a question: Why are many people convinced that there is no such thing as a false fact, when hardly anyone denies the existence of false teeth, false bottoms, and false killer whales? For comparison, let's take the semantically similar word fake. Something that's fake appears to be the actual item, but isn't. Substitute fake for false in false teeth, false bottoms, and false killer whales, and the meanings don't change, although people might remark upon your slightly unidiomatic usage.
How about a fake statement? That would be something that appeared to be a statement, but was actually something else. I can't quite imagine what such a thing would be, but such is the meaning of fake that it has caused me to try to imagine it. In contrast, I can easily get a meaning for fake fact. It would be something that appears to be a fact but isn't, such as a made-up detail in a backstory created for someone in witness protection.
Now let's reconsider false. Although in many respects it's a synonym for fake, it doesn't behave quite the same. Let's take a few collocations with fake — fake ID, fake fur, and fake blood — and replace fake with false. Again, the meanings don't change, so up to this point, false and fake are interchangeable. However, we do get a clear difference for false statement. This noun does not have the hard-to-imagine meaning of fake statement; it simply means a statement that isn't true. In other words, when false modifies a noun that denotes some kind of proposition, it means that that proposition isn't true. Similarly, a false claim is not something that appears to be a claim but is really something else. (What would such a thing be, anyway?) It just means that the proposition denoted by the claim isn't true.
And in just the same way, when false modifies fact, it doesn't mean something that appears to be a fact but isn't, like the details in a cover identity for a spy. It simply means a fact that is not true — a meaning that requires fact to refer to a proposition that might or might not be true. The alt.usage.english defense would be valid if false fact meant the same thing as fake fact. As it is, the argument overlooks the different meanings of false when it modifies ordinary nouns and when it modifies nouns that refer to some kind of proposition.
Of course, even if we reject the logically flawed defense of true fact and false fact, there's still the usual defense that serves for decimate, awful, and all the old favorites: Words and meanings change over time, and this word has had this meaning for many years.