Dismantling Communication—Literally

Phobia by Allie Holzman

A friend was telling me about a particularly awkward experience recently—which I will not share here, except to remind you that coffee is indeed a diuretic—when she happened to say, “I literally died of embarrassment.”

It was hard not to offer a snarky reply. Literally died? What, exactly, was she asserting? “I was murdered by my own shame. And, on the third day, I rose again so as not to miss this coffee date.”

I held my tongue: honestly, she had been through enough. Except, of course, a literal death. It would have been more truthful to say, “I figuratively died of embarrassment.” Or, “It was like I died of embarrassment.” But literally? She hadn’t died. She’d lied.

And it’s a dangerous thing, to say things that aren’t true—even when no one is deceived. We expect certain behaviors of one another in conversation, and these shared assumptions enable effective communication. Paul Grice outlines these expectations with four maxims that linguists refer to as “the cooperative principle”:

  1. Quantity: Be as informative as required—not more or less.
  2. Quality: Don’t say things you believe to be false, or don’t have any evidence to support.
  3. Relation: Be relevant.
  4. Manner: Be lucid, avoiding ambiguity and wordiness.

Communication breaks down if we do not follow these maxims most of the time. Consider Quantity. “I have one brother,” I might say. And this is true—but I have another brother as well. If you couldn’t count on me to follow the Maxim of Quantity, you’d have to check every assertion I made. Our conversations would become laborious and unpleasant.

Or say I’m going for a walk. My roommate, looking at the Paris weather report, announces, “It’s raining.” But (as I discover after donning my boots and raincoat), this is irrelevant to the walk I’m taking in Boston. If I can’t count on my roommate to follow the Maxim of Relation, it becomes much more difficult for us to communicate. Grice’s Maxims form a common foundation that allows us to interpret and understand one another. So by asserting—despite knowing it to be untrue—that she literally died, my friend deliberately chipped away at the Maxim of Quality.

Her use of “literally” violates the Maxim of Quantity as well. “I died of embarrassment,” she could have said. Adding “literally” to her statement doesn’t add any new information. The words she used already had their literal meaning. She’s already said, literally, that she died of embarrassment.

So why say it? Why violate two maxims of communication, maxims that make it possible for us to understand each other? Our ability to communicate is at stake. Why risk it?

After all, thanks to the Cooperative Principle, we do manage to communicate successfully most of the time. The salami is in the refrigerator, and the cheetah is the fastest mammal: not much question there. We make certain assumptions, and usually it works out just fine. But usually—we say things people expect to hear, don’t we? Maybe no one has said this particular sentence before. But for the most part, we’re really talking about things that have already been talked about for years and years and years.

Some of it is new, of course, and that’s where we run into trouble—trying to explain a new way to solve a problem, or a new way to look at the world. This is why literature is often so complicated: it gives us new ways of seeing. In his book Principles of Art, R.G. Collingwood describes literary expression as “like a gull over a ship’s stern. Trying to fix the proper meaning […] is like coaxing the gull to settle in the rigging.” Literature is trying to say something new—which means we don’t have words for it yet, although of course words are all we have to work with. We have to coax the gull to settle. Or, in the angstier promulgation of Eliot’s J. Alfred Prufrock, “It is impossible to say just what I mean!”

When I first started studying poetry, I used to wonder: why do poets insist on obfuscating the thesis? But it is impossible to say just what I mean! There are no words for the thesis yet. Poets have to do tricky things with the words we already have, in order to get at the things we’ve never talked about. The first jilted lover finally settled on the expression “broken-hearted” to help us understand her pain, and it worked well enough for a while—and still does work, in fact, if all you want to express is that your pain is the same as that of all the jilted lovers we’ve seen since then. (And let’s be honest, it probably is.) But if you want us to think about this particular heart, this particular pain—we find ourselves back with Collingwood, coaxing a gull onto a rigging where it has never perched before.

When people use the word “literally” before a metaphor, it’s tempting to respond: “You don’t really mean that literally. You mean it figuratively.” But the fact is, we always mean our words figuratively. There’s nothing inherently salami-like about the word “salami”. We connect them through our shared assumptions, our common uses, our convention that the word will be employed in particular ways. Every word begins as a metaphor. Rather than say “I literally died of embarrassment,” perhaps it would have been more truthful for my friend to say, “I was mortified.” But is that so different, in the end? After all, the root of the word means “to put to death.” It, too, carries the relic of figurative meaning.

“Literally” reminds us of the medium we use to communicate. No matter what we say, we will have said it literally—simply by virtue of speaking with words. But explicitly flagging speech as literal calls attention to a fact often overlooked: I used language to say this, and it is impossible to say just what I mean. Or, put another way, sometimes we mean something different from what our words mean. Saying “literally” reminds us that metaphor is an option, too.

We’ve become inured to most of the metaphors we encounter on a day-to-day basis. People admire the blanket of snow or lament a broken heart so often that we forget that these are metaphors. And in some sense, they aren’t metaphors anymore. For most of us, “a broken heart” means roughly the same thing as “emotional distress caused by a loved one.”

But saying “literally” breathes new life into the metaphor. It reminds us to look at what the words themselves are actually saying. “Broken-hearted” is a tired epithet. But “literally broken-hearted” is much more interesting. Of course the heart in question did not split apart. That was true of “broken-hearted” and it is still true of “literally broken-hearted.” But “literally” says, listen to the words I’m using: I mean them! So we listen. And we think about what a literal broken heart might have in common with emotional distress. We imagine the pain in the chest and the feeling that the body cannot survive this way, rent in two.

Metaphor uses blatant untruth to pull us out of our everyday world. “But soft! What light from yonder window breaks?” Romeo asks. “It is the East, and Juliet is the sun!” Romeo has not gone insane when he says this. He’s madly in love but he isn’t mad. Nevertheless, in our world Juliet certainly is not the sun. Romeo is asking us to join him in another world, where she is the sun, to see how things look. Denis Donoghue writes about the “impulse in metaphor to escape from the world.” We might add that, to do so, a metaphor-maker risks the very foundation of our communication. Language breaks down if we do not say true things most of the time. There must be something very important about Juliet that Romeo wants us to see, that he would stake on it communication itself.

There’s a Calvin and Hobbes cartoon where Calvin’s father is quietly reading his book. Calvin stands right next to him, inflates a balloon, and pops it. Of course his father is horribly startled. Calvin turns to him and says, “Pay attention to me.”

That’s what “literally” does. When I use a metaphor, don’t just say, “That’s nice, dear,” and go about your normal life. Pay attention to me! There’s something here I actually want you to think about! The word “literally” is my way of popping a balloon by your ear: listen to the words I’m using. Join me in the world I’m proposing.

We need common assumptions in order to communicate. We need to be able to count on each other for quantity, quality, relation, and manner; just as surely as we count on a common vocabulary as English speakers. Every violation of these conversational maxims diminishes the shared foundation that lets us communicate. Every time we say “literally” when “figuratively” would be closer to the truth, we disrupt those common assumptions. And for what? What do we gain by making it harder to understand each other? What is the value in rendering language unreliable?

In fact, language is already unreliable. We forget this amid ostensibly concrete expressions and metaphor that has long since turned to idiom. But “literally” reminds us: we can use words in many different ways; and we shape their meaning with every use. Thus “literally” awakens us to the power of the unreliable. This is how new things get said—through metaphor, through the uncommon use of common words, through the coaxing of the gull.

“I literally died of embarrassment,” my friend said. It’s not true, and such untruths could make everything come undone. But language is already undone. It’s up to us to find new ways to tie it together.


Works cited in the text:

Collingwood, R.G. The Principles of Art. Oxford University Press, USA, 1958; p. 7
Donoghue, Denis. Metaphor. Cambridge: Harvard UP, 2014; p. 204
Eliot, T.S. “The Love Song of J. Alfred Prufrock.” Literature: Reading Fiction, Poetry, and Drama. Ed. Robert DiYanni. 6th ed.
Grice, Paul. “Logic and conversation.” 1975.  In Cole, P., and J.L. Morgan, eds. Speech Acts. New York: Academic Press, 41–58
Shakespeare, William, and G. Blakemore Evans. Romeo and Juliet. Cambridge, UK: Cambridge UP, 2003; Act II Scene 2


Photo “Phobia” by Allie Holzman ; licensed under CC BY 2.0

Maisie Wiltshire-Gordon
Latest posts by Maisie Wiltshire-Gordon (see all)


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.