We need to talk about literally
Hold the front pages, literally. Or not. There has been much excitement this week over the discovery that the Oxford English Dictionary (OED) has recorded a sense of the word literally that seems to cause particular irritation. I am speaking of its use in a sentence like “I literally died laughing and had to run out of the room before I disrupted the meeting”. Of course, had the writer actually died, then the meeting was probably already disrupted, not least by the sight of a corpse running out of a room.
I jest, but it is precisely this type of disputed usage that has people riled, and when they find that it is included in Oxford Dictionaries, they can react as if the world is literally coming to an end. What may surprise them is to find that this usage is much older than you would think. While it is true that it has become increasingly common in modern usage, it was actually first included in the OED in 1903. When the entry was updated and published online in September 2011, we found even earlier examples of this usage – our earliest example is currently from 1769.
This newer, disputed usage (describing something non-literal, as a form of exaggeration) has become more frequent over time, and is now sometimes used quite deliberately in non-literal contexts. But literally has always been employed for added effect or emphasis. If you write, “the book is literally 500 pages”, the statement may be precise, but its precision doesn’t depend on the presence of the word literally; indeed the emphasis it adds may introduce doubt about the precision.
Why does this seem to be so much of a problem? Is it to do with logic? “I literally died laughing and had to run out of the room before I disrupted the meeting” couldn’t possibly be true, logically speaking, and for some language should be logical. Take double negatives. Sentences like “I didn’t do nothing” often meet with objection – “well, if you didn’t do nothing, you must have done something”. But while multiplying two negatives in mathematics certainly make a positive, language works in more mysterious ways. Pick up a copy of the Canterbury Tales by Chaucer and you will find numerous examples of double, triple, or even quadruple negatives. In Middle English, the use of more than one negative served to intensify the force of the negative being expressed. In the Prologue to the Canterbury Tales, Chaucer wrote of the very perfect and gentle Knight: “He neuere yet no vileynye ne sayde In al his lyf vn to no maner wight”. (Never in all his life had he spoken rudely to anyone). Something similar is going on in modern English. If someone says “I didn’t do nothing” nowadays, he or she is usually emphasizing that nothing went on or nothing happened. Nowadays, this usage is not one which you would expect to find in formal, especially written contexts, but it is especially common in dialect and rarely (if ever) causes genuine confusion, and can often give a subtle shade of meaning. Double negatives can also form a nuanced way of expressing a positive, but the meaning comes from the context or manner of expression. “I didn’t do nothing” can mean “I did the bare minimum” and “I didn’t do nothing” can mean “I did plenty”.
Confused? You won’t be
It is also unlikely that someone would genuinely be confused by the sentence “Rooney was literally on fire”. How about “I was literally coughing my guts up”? Perhaps part of the problem lies with what literally is qualifying in these examples, i.e. phrases that themselves are figurative, or dare I say it, not meant to be taken literally (in one of its senses). So, by adding literally to such expressions, you are reversing the figurative nature of them. Consider the sentence “He has literally put blood, sweat, and tears into earning a living for his family, but it goes with the turf”. If you removed literally from this sentence, how does the meaning change? It doesn’t in essence. The exaggeration is lost, certainly, but “put blood, sweat, and tears” is still figurative. What if you put “truly” or “really”? Would that cause as much consternation? It still wouldn’t make the sentence literal, but I suspect it wouldn’t cause quite so much opprobrium.
Something wicked this way comes? Nice
Of course the use of literally in these examples is seen as a reversal of the original meaning, and that might be key to the problem. Yet a word having opposite or even multiple meanings in existence at the same time is hardly unusual. There are plenty of examples of words which have opposing meanings (and some in which you might genuinely need context in order to make sense of things). Wicked can mean “evil” or it can also mean “impressive” (since at least 1920); sick can mean “offensively unpleasant” or it can mean “excellent”; nonplussed means “surprised; confused” and in recent years has developed a meaning of “unperturbed”. Admittedly, these senses are not ones which would be used in formal contexts (all are marked as informal in our online dictionary). So how about the word “nice”? The earliest sense of nice as recorded in the OED means “foolish “or “ignorant”. At the same time, it meant “wanton” and also “refined”. So the phrase “she’s a nice girl” taken out of context could cause confusion. When Shakespeare wrote of “nice wenches” in Loves Labours Lost, he meant “wanton”, yet both of those other senses were available (as well as others).
Words can have more than one meaning, sometimes at the same time. This isn’t a new concept. Perhaps the problem with a word like literally is that we are looking at it in a vacuum, as the word exists now. With nice, we don’t have any difficulty with it having opposing meanings because one primary meaning has taken over and many of those others are long obsolete (of the 42 senses of the adjective nice included in the OED, 22 of them are now obsolete). We don’t come across them that often. But that wouldn’t always have been the case.
Whatever the reasons, it is clear that people often have strong opinions about “new” senses of words. Perhaps the question is not so much why do people have a problem with literally but rather why do lexicographers not have a problem? It comes down to that oft-spoke mantra – language changes. Our job is to document that for better or for worse. Except for us, there is no worse. We have to look at language objectively and dispassionately. Of course, part of our job is to give guidance on what might be acceptable when. That is why we label some words as slang and why we give a usage note at the offending sense of literally, making clear that although it is very common, it is considered irregular in standard English.
Which is why we literally cannot see what all the fuss is about.