Metaphor Creep
The sneaky side effect of comparisons
Consider the following pair of sentences:
His body was shaped like a pear, with narrow, sloping shoulders and a heavy gut.
His body was shaped like a bowling pin, with narrow, sloping shoulders and a heavy gut.
Between the two possibilities, 1 and 2, which is better?
Certainly, someone could make the case that 1 has the virtue of being marginally more concise, 2 marginally more original. I suppose it's also true that a pear is not exactly the same shape as a bowling pin, and one option might serve the writer’s need better for the sole reason of physical resemblance. Overall, though, I hope these sentences feel equivalent to you but for the (nearly interchangeable) objects to which the man’s body is compared.
Or do they…? While the similes purport to describe shape alone, does each not also bear the subtle imprint of two men whose lives, on closer examination, have less in common than their similarly chubby tummies might suggest? For me, it’s hard not to imagine Pear Man as a dumpy, sweetly inoffensive guy. Bowling Pin Man seems stiffer, unstable even—a person liable to get knocked around by the hard spin life sends his way. Look back at the sentences; do you not feel the ever-so-slightly different characterizations tugging at your consciousness?
I’m stretching these metaphors pretty thin, but I hope you can see my point: In addition to general principles of good writing, the word choice here should be dictated by the implicit associations each object will bring to the reader’s mind. Together, the two sentences are meant to illustrate an idea I've been playing around with, a phenomenon I’m going to propose we call metaphor creep. Here’s a definition: metaphor creep—the tendency of analogies to expand beyond the similarity they first seem to highlight.
Taken seriously, metaphor creep implies that the comparisons we make can shape perception in hidden or unexpected ways, and that skillful communication requires careful attention to—and, often, deliberate harnessing of—these implicit associations.
A Note on Terminology
One irony of deciding what to title this post is that the phrase metaphor creep is itself subject to metaphor creep—i.e., in deciding what to name the concept, I am inevitably committing myself (and you!) to a certain framework for thinking about it.
Originally, for example, I had formulated the idea as analogic contamination, but I didn’t like the implication of disease or dirtiness, and worried it sounded obnoxiously academic. Metaphor drift and metaphor seep felt closer to what I was going for—but “drift” seemed to suggest unidirectional movement, whereas I had in my mind’s eye a meandering, less predictable spread; “seep” offered almost the opposite problem, seeming too uncontrollable and directionless to fit the concept. (Among other rejected terms: metaphor runoff; metaphor spillage; metaphor diffusion; metaphor aura; metaphor residue; metaphor expansion; metaphor contagion; metaphor radiation; metaphor bloat; metaphor projection; and metaphor liquidity. I’m sure there are other preexisting terms that smart people have come up with for the same, or similar, concepts, too.)
Ultimately, I settled on metaphor creep for its evocation of vinelike growth, slight uncanniness, and sneakiness. As I will explore, these qualities can be leveraged toward both positive ends (as in subtly referential writing) and more sinister ones (as in dog-whistley political rhetoric).
I should also note that I'm using “metaphor” here loosely, to refer to any number of literary devices that liken one thing to another. This includes extended comparisons (i.e., analogies and allegories) and various classes of figurative language (e.g., similes, personification, hyperbole, idioms, etc.).
The Power of Connotation (a.k.a. Language Creep)
Everything I’m arguing here about metaphors is also true, to a lesser extent, of individual words. I just happen to believe that metaphors serve as a particularly potent influence on our perceptions, and that they make for particularly fruitful and entertaining examples! In its most reduced form, though, this post is an argument to simply pay closer attention to the power of connotation in general, the tendency of all language to evoke and expand in our minds.
What is actually happening inside our heads when language creep is occuring?
In his recent New Yorker article, James Somers describes the efforts of early-20th-Century psychologist Charles Osgood to measure the subtle semantic distinctions our brains make between words:
Osgood conducted an experiment. He asked people to rate twenty concepts on fifty different scales. The concepts ranged widely: boulder, me, tornado, mother. So did the scales, which were defined by opposites: fair-unfair, hot-cold, fragrant-foul. Some ratings were difficult: is a tornado fragrant or foul? But the idea was that the method would reveal fine and even elusive shades of similarity and difference among concepts. “Most English-speaking Americans feel that there is a difference, somehow, between ‘good’ and ‘nice’ but find it difficult to explain,” Osgood wrote. His surveys found that, at least for nineteen-fifties college students, the two concepts overlapped much of the time. They diverged for nouns that had a male or female slant. Mother might be rated nice but not good, and cop vice versa. Osgood concluded that “good” was “somewhat stronger, rougher, more angular, and larger” than “nice.”
Osgood eventually came up with three conceptual axes to categorize words (one for pleasantness, one for potency, one for passivity) based on patterns in subjects’ responses. Then in the 1980s, other researchers built on his methodology to identify “a few hundred dimensions,” many of which “described abstract or ‘latent’ qualities that the words had in common—connections that wouldn’t be apparent to most English speakers. The researchers called their technique ‘latent semantic analysis,’ or L.S.A.”
In practice, I take L.S.A. to mean that a speaker/writer can use connotations to evoke certain topics without explicitly naming them. Of course, this is easy to do using a word with very particular associations. (The minute a news anchor calls someone a “snowflake,” you have a pretty good idea of the kind of segment you’re about to watch; ditto for a book/film that mentions a vampire, zombie, or other genre-specific word.) Most common words, however, bring to mind a vast array of associations, each one with unrelated or even conflicting connotations of its own—like verbal lanterns that, in isolation, cast their semantic rays in countless different directions.
Take the word nail, for example. For me, it brings to mind all of the following: hammers, metal, salons, fingers, claws, “tooth and nail” and other nail-related expressions—and even words bearing a close phonological resemblance, like mail. (And these are only the associations that I’m conscious of, that I’ve been able to generate off the top of my head!)
Put multiple words in a more circular arrangement, however, and their light begins to converge on a common center—even when that center is never mentioned overtly. If, in addition to nail, I give you wart, cat, and broom, for example…
Things are suddenly beginning to feel pretty witchy, aren’t they?
Note that in order to achieve this effect, it need not be the case that any one word bear extremely close relation to the unnamed subject. In fact, with the possible exception of wart, I doubt witch would be the first word that comes to mind for any of them; for nail it might not even crack the top fifty (hence the dotted line).
In the context of the other three words, however, nail does lend one more piece of evidence, however weak, to the presence of a witch in the mix—or at least it logically fits with this interpretation whereas not all words would. If, for example, we are introduced to a fictional cat owner with a penchant for sweeping, we may find her odd but are unlikely to suspect knowledge of spells and potions. If these two details are mentioned repeatedly, however, and the writer goes on to describe the woman’s warts and unusually long nails, suddenly the accumulation of images begins to take on a slightly eerie significance. (Of course, even this might not be enough for the truth to percolate all the way up to conscious thought. Give her skin a green tint and put a bubbling cauldron in her kitchen, though, and our alarm bells will most certainly start ringing—all without having once read the word witch.)
Again, there is not one clear metaphor in this example, in the sense of “cat = X, broom = Y”—and hence no metaphor creep per se. This is just an illustration of the idea that when readers’ associations are predictable, a writer may be able to use them to subtly direct attention. Language (or, really, the way our minds interpret language) is inherently creep-y.
Unintended Creep
Unfortunately, readers’ associations are not always predictable.
Inevitably, different audiences will bring their own sets of idiosyncratic responses, of which the author cannot possibly be completely aware in advance. Here’s Thomas Lux on this point, in a lovely excerpt from his poem “The Voice You Hear When You Read Silently”:
It is your voice
saying, for example, the word “barn”
that the writer wrote
but the “barn” you say
is a barn you know or knew. The voice
in your head, speaking as you read,
never says anything neutrally—some people
hated the barn they knew,
some people love the barn they know
so you hear the word loaded
and a sensory constellation
is lit: horse-gnawed stalls,
hayloft, black heat tape wrapping
a water pipe, a slippery
spilled chirr of oats from a split sack,
the bony, filthy haunches of cows...
And “barn” is only a noun—no verb
or subject has entered into the sentence yet!
As Lux says, mere words evoke reader responses, unbidden; metaphors are effective because they actively invite one’s audience to find an implied connection. Unfortunately, this means they may also invite people to infer meaning that is not necessarily intended
Let me show you what I mean by way of an example from Shrek:
(Here’s the key dialogue if you can’t spare ninety seconds, or your boss is around or whatever:
SHREK: Ogres are like onions.
DONKEY: They stink?
SHREK: Yes—no! [correcting himself]
DONKEY: Oh, they make you cry.
SHREK: No.
DONKEY: Oh, you leave them out in the sun, they get all brown, start sprouting little white hairs?
SHREK: No! Layers. Onions have layers. Ogres have layers. Onions have layers—you get it: We both have layers.)
The comedy, of course, stems from Donkey’s initial misinterpretation of Shrek’s metaphor—but his confusion also highlights how little control a metaphor’s author has over how it is received: The mind creeps where it will, similarities speak for themselves. In fact, I’d argue that the whole scene depends on the audience agreeing that ogres possess an essential oniony-ness that extends beyond the layer similarity. (Shrek himself even inadvertently affirms Donkey’s stink point.) Ogres are like onions because both have layers, yes—but also because ~vibes~. Textbook metaphor creep.
The second half of the clip in particular drives this point home. Here, Donkey suggests that Shrek could just as easily compared himself to a cake or a parfait as to an onion. Of course, this suggestion feels…not right—which is why it also plays as a joke. But why is Donkey wrong exactly? Sure, the desserts have vertical layers while Shrek’s simile is suggestive of interiority/exteriority, but in both cases, you have to work your way through the top part to get to what’s underneath; it would still basically communicate Shrek’s message, right? Yet it feels wrong on a gut level. It’s almost like there’s a Stroop effect going on here, dare I analogize, where our brains are unable to separate out the relevant information (i.e., layers) from other mostly irrelevant factors (i.e., sweetness, appearance, and so on).
Except metaphor creep isn’t a bug; it’s a feature of how our brains work. As Shrek himself says, “I don’t care what everyone likes—ogres are not like cakes…Ogres are like onions. End of story.” We are capable of detecting deep resonances even when we are not capable of articulating strict reasons for them.
How can people take advantage of this fact to communicate more effectively?
Crafting Effective Metaphors
Unintended associations are just one reason to deploy metaphors with caution—per Elements of Style:
The simile is a common device and a useful one, but similes coming in rapid fire, one right on top of another, are more distracting than illuminating. Readers need time to catch their breath; they can’t be expected to compare everything with something else, with no relief in sight.
When you use a metaphor, do not mix it up. That is, don’t start by calling something a swordfish and end by calling it an hourglass. (Strunk and White, 80).
(I’d make the case that the real problem with mixed metaphors is that they result in competing metaphor creeps: The swordfish calls to mind one set of associations, the hourglass another, and the clash reduces the coherence of the overall picture.)
What, then, makes for an effective metaphor?
Although this is more of an art than a science, writer Ocean Vuong suggests the following two criteria for achieving maximal evocative power:
1. [A] sensory (visual, texture, sound etc) connector between origin image and the transforming image.
2. A clear logical connector between both images.
(In the terms of my original metaphor creep definition, #1 ≈ the highlighted similarity and #2 ≈ the metaphor’s implicit expansion.)
#1 is the easy part: If you want to emphasize how great or bright or [your adjective here] something is, the comparison either works or it doesn’t. If you want to nail #2, though, you have to trust your gut to detect latent resonances—and, even more difficult, whether they will resonate with your audience as well. Though Vuong is referring mostly to poetry, I think his rules hold true in a broader social/cultural context as well, as I will explore in the section that follows.
Is it possible for a metaphor to have too much resonance/too much of a logical connector? I think so. Consider these examples:
The winter wind was as cold as ice.
I felt like a million bucks when I finally got my first paycheck.
Trying to get around without a car is no walk in the park!
The problem with these analogies, other than reliance on overused language, is that the tenor and the vehicle in each are so closely related that they naturally creep into one another: “Winter wind” brings to mind “ice” by near-automatic association, as “paycheck” does “bucks.” Example C is perhaps less egregious (it has a bit of a Yogi Berra quality, which could be desirable in certain situations?), but note that it’s essentially comparing the difficulty of walking to…a difficult walk.
It’s no coincidence that these examples all include common figures of speech, either. As George Orwell argues in “Politics and the English Language” (see “Dying Metaphors”), it is precisely because these cliches have entered into the culture as easy short-hands that they fail to evoke much on the part of the audience at all. We don’t pause to actually imagine the metaphors, in other words, because we feel we already know them. If I had given an original bad simile, as opposed to an overused one, (e.g., The tree was as green as a field of springtime clovers), I expect the tautology would jump out even more.
What about the opposite problem—i.e., deliberately pairing two objects/images whose logical relation is not obvious?
When done poorly, this creates absurdity (The tree was as green as a chair)—but if it’s pulled off, the result can be delightfully poetic or provocative. In the found poem, “One Boy Told Me,” for example, Naomi Shihab Nye compiles such formulations from her young son as, “Your head is a souvenir,” “Oatmeal cookies make my throat gallop,” and “My tongue is the car wash / for the spoon.” Despite the low likelihood of any of these metaphors occurring to most adults, they are memorable and impactful. For me, most of them work on some level, either by inviting particularly imaginative metaphor creep or tapping into some difficult-to-articulate resonance.
A skillful author can also harness a metaphor’s internal associations—i.e., those that occur while in the midst of reading it, as opposed to those that emerge afterwards. This is especially true in extended metaphors, whose multiple sentences create space to play with reader expectations.
Our pattern-finding brains, after all, are making constant and automatic predictions about the world—including about the language we receive. You may recall that near the top of this essay I suggested that words could be arranged like lanterns to cast their light on an unnamed center (e.g., nail + wart + cat + broom = witch). Put words in a syntactic arrangement instead, like hit the nail on the…, and our minds begin to creep out ahead of the unfolding sentence: The light of words illuminates a forward path—a row of streetlamps leading the reader toward an implied destination.
In support of “internal creep,” I’ll close this section by looking at two extended metaphors from beloved books—Philip Pullman’s YA classic The Golden Compass, and Toni Morrison’s moving second novel, Sula. Here’s Pullman:
Then, with a roar and a blur of snow both bears moved at the same moment. Like two great masses of rock balanced on adjoining peaks and shaken loose by an earthquake, which bound down the mountainsides gathering speed, leaping over crevasses and knocking trees into splinters, until they crash into each other so hard that both are smashed to powder and flying chips of stone: that was how the two bears came together. (350)
Initially—for me at least—the “two great masses of rock” are disorienting: Though we read them as somehow referential to the two bears, their exact significance remains ambiguous, and we search desperately for the similarity they are intended to highlight. As the rocks are described moving downhill, however, our associative minds begin to anticipate the violent collision before it actually occurs, creating suspense and making the words all the more satisfying when they arrive. This, I assume, is a major source of its over-the-top, epic-sounding quality. (You may even remember learning that this is called a Homeric simile when you were forced to read The Iliad or The Odyssey in high school.)
What would a more complex shape look like in an extended metaphor—one where the imagery isn’t circling around an implied center but not quite obeying a single linear logic either?
Toni Morrison offers one possible answer in Sula. In the passage that follows, one of the main characters, Nel, has just discovered [SPOILER ALERT] that her husband and lifelong best friend have been engaged in an affair. Upon the revelation, Morrison writes, of Nel:
She stood up frightened. There was something just to the right of her, in the air, just out of view. She could not see it, but she knew exactly what it looked like. A gray ball hovering just there. Just there. To the right. Quiet, gray, dirty. A ball of muddy strings, but without weight, fluffy but terrible in its malevolence. She knew she could not look, so she closed her eyes and crept past it out of the bathroom, shutting the door behind her. (109)
This is a notable example for a few reasons. Metaphors for emotion abound in English, of course (I’m feeling blue; That’s a huge weight off my shoulders), and Morrison is hardly the first person to embody/concretize mental illness (e.g., the black dog, the Babadook). Here, though, we get a comparison not to an easily recognizable image but to an invented object, almost a nonsensical one. This feels like it shouldn’t work, so why does it?
Again, I think the answer has a lot to do with internal metaphor creep, specifically the way Morrison uses suspense to set up the gray ball. I don’t know about you, but reading the first half of the quote (“There was something…just out of view. She could not see it, but she knew exactly what it looked like”), my mind actually moves to something much closer to the anthropomorphic examples linked in the previous paragraph. It reads like the setup for a jump-scare, in other words: the tense, quiet moment before the monster appears on-screen for the first time.
The effect of these sentences, and context of the plot around them, is to imbue the concrete image that follows with a lot of emotional content that we would not otherwise associate with the gray ball. It’s a made-up object, after all, and therefore inherently strange and surprising—but as we perceive it in our mind’s eye, the dread from the setup lingers, lending coherence to its otherwise disparate components. Suddenly the metaphor becomes unsettling, and ripe with potential meanings.
(Note also that the gray ball sequence is the opposite of the way metaphors typically work. In Ogres are like onions, or the Pullman excerpt, we get the concrete comparison first and then build out abstractions and other meanings from it second. In Morrison’s case, it’s mostly the why that comes first and the what that follows—though of course we can still extrapolate further meaning from the gray ball after hearing it described.)
In short: Metaphors are inherently squishy and tough to pull off, so it’s hard to come up with any universal prescriptive rules for them. Probably the best an author can do is try to anticipate a given metaphor’s full range of connotations, and do their best to hit the sweet spot between clear connection and surprising transformation.
So What? (Why This Matters in the Real World™)
The fundamental lesson of metaphor creep is to simply pay greater attention to the kinds of analogies we choose to draw.
In fact, a sufficiently creative person in search of a suitable comparison will usually have plenty to choose from: A difficult project has, in the language of everyday metaphors, a lot of moving parts or requires keeping a lot of balls in the air; it may be an uphill battle, a pain in the ass, a Gordian knot, a beast.
While these various comparisons all make more-or-less the same explicit point, they do, on closer inspection, imply different psychologies and courses of action. (Does the project require routinization of systems or skillful prioritization by individuals? Is success the result of persistence, resignation, a radical reframe, or mastery over one’s enemies?) Even when these distinctions feel trivial, as with the pear/bowling pin example at the start of this essay, they will still result in slightly different flavors of sentence. Compound those distinctions over the course of many words and paragraphs—or, alternately, debates, meetings, policies, etc.—and the differences can become more substantial.
The challenge, in many cases, is that the effect on our thinking is organic and difficult to detect—i.e., creep-y—as it grows out of an initial choice that we may not even have been aware we were making. Here’s Annie Murphy Paul, for example, in her 2021 book The Extended Mind, on our dominant analogies for the brain:
These two metaphors—brain as computer and brain as muscle—share some key assumptions. To wit: the mind is a discrete thing that is sealed in the skull; this discrete thing determines how well people are able to think; this thing has stable properties that can easily be measured, compared, and ranked. Such assumptions feel comfortably familiar; indeed, they weren’t particularly novel even at the moment they were first proposed. For centuries, brains had been likened to machines—to whichever appliance of the time appeared most advanced: a hydraulic pump, a mechanical clock, a steam engine, a telegraph machine. (9)
She goes on to propose an alternative metaphor for the brain, the magpie, which offers “very different implications for how mental processes operate. For one thing: thought happens not only inside the skull but in the world, too; it’s an act of continuous assembly and reassembly that draws on resources external to the brain” (11). This is more or less the thesis of her book, and it guides her to discuss mind extensions as wide-ranging as nature, modern technology, and hand gestures. (I recommend!)
Whether Paul’s magpie analogy is guilty of its own unintended consequences, I can’t say; it’s certainly a memorable image that helps hold the book together. It would be unfair, too, I think, to reject the standard comparisons without also providing one of her own. Without analogies, after all, how would we make meaning of the world in the first place? What tools do we have to understand the abstract or unfamiliar, really, beyond those concepts of which we already have a grasp?
In both policy and in politics, the way we frame problems has a profound impact on the set of solutions we entertain. If a coalition succeeds in establishing a dominant frame for an issue, therefore (almost always pushing an implicit or explicit metaphor in the process), we all reap the costs and benefits of this frame’s implications—i.e., its metaphor creep.
Take the now infamous war on drugs, for example. I’ll caveat the claim I’m about to make by noting that it’s easy to declare outcomes obvious in hindsight, and that there’s no way to run the counterfactual here…Still, does it not seem almost inevitable that such an overt and relentless militaristic framework would lead to an overreliance on state violence to achieve its ends? (In his 1971 speech declaring drug abuse “Public Enemy Number One,” President Nixon repeatedly likens his efforts to a literal war mobilization, referring to the initiative as an “offensive” seven times in the span of less than five minutes.) Whether or not Nixon and his successors were motivated solely by racial animus, a “war” frame gives rise to extreme, binary thinking; it feeds a need for heroes and villains, encouraging negative stereotypes and the coded language used to activate them (e.g., thug, welfare queen, law and order).
Again, I don’t want to be so reductive as to suggest that police brutality and mass incarceration can be explained simply by the presence of militaristic language. But surely rhetoric has been a necessary, if not sufficient, factor in America’s collective response to drugs. What might the country look like if an economic, ecological, or public health framework had captured the public imagination instead?
While metaphor creep can certainly be weaponized (if the above example is too speculative for you, consider blatant euphemisms like ethnic cleansing or enhanced interrogation), political metaphors are neither inherently nefarious nor associated with any one movement/party.
In A Letter from Birmingham Jail, for example, Martin Luther King Jr. offers this striking analogy: “For years now I have heard the word ‘wait.’ It rings in the ear of every Negro with a piercing familiarity. This ‘wait’ has almost always meant ‘never.’ It has been a tranquilizing thalidomide, relieving the emotional stress for a moment, only to give birth to an ill-formed infant of frustration” (2). (Emphasis mine; for those, like me, too young to have much direct awareness of thalidomide, it was a nausea-reducing drug given to pregnant women that was later discovered to cause severe birth defects.)
Again, there are any number of comparisons one could make to express the basic idea here—something like: short-term “solution” that creates or exacerbates a more significant down-stream problem. Off the top of my head, alcoholism, reckless spending, and fossil fuel consumption would all be shittier but “correct” substitutes for thalidomide, as would the preferred choice of today’s politicians: a band-aid. The inferiority of these versions is obvious, I hope, for their lack of both memorableness and persuasive force. (See Orwell point about “dying metaphors” in previous section.)
In particular, what makes MLK’s choice so effective, I think, is the visceral specificity of “ill-formed infant.” The moral urgency this image provokes is the result of metaphor creep: He doesn’t need to tell us how wrong the “wait” is because our minds already go there. Rhetorically, this forces a debate on King’s terms, placing the burden of proof on detractors to show why delaying justice is unlike thalidomide if they don’t like the metaphor’s implications. (Of course, I doubt many segregationists were engaging with him on the substance, but surely his words found their way to at least a few of the white moderates he name-checks later in the letter.)
Metaphors are an essential communication tool for anyone seeking to influence public perception. How many times have you heard a political candidate talk about growing the pie or helping individuals to climb the socioeconomic ladder? How many times have you heard welfare spending described as either an investment/human infrastructure or as the purview of a nanny state? All of these metaphors do arguably capture something about the function of government, but they push against our intuitions in conflicting ways to make different issues feel more or less salient. (Yes, maybe we can walk and chew gum at the same time in tackling these issues—or maybe, the proverbial gum budget and walking budget are actually in competition for a limited pool of funding and political capital.)
In short, dominant metaphors have real-world impacts on the people and forces governing the way we live. The most insidious ones can shape understanding of an issue without our even noticing; the most persuasive ones can prompt an epiphany, shedding sudden light on a perspective that would never before have occurred to us.
TL;DR
Because of the way our minds process language, metaphors have a tendency to expand outward, or “creep.” This is true to a lesser extent of individual words.
For this reason, it is important to craft metaphors with careful attention to their latent implications, especially since they may mean different things to different people. Great writers/communicators/persuaders use metaphor creep deliberately to create analogies that both surprise and, on an intuitive level, work.
If you’re someone with a significant platform/influence, like a journalist or a politician, be aware!
Thanks to Charlie, Ari, and Rachel for edits.
Housekeeping/FAQS
What is this newsletter about?
Whatever I want/am interested in—which will probably mean a lot about writing, education, psychology, politics, games, and such. I imagine this first post is fairly representative. I may also put up fiction or other creative projects that don’t have a home elsewhere.
How frequent will posts be?
My goal is to put something out 1-3 times/month.
Who are you?
I’m a writer and former high school English teacher. David Gamson is a pseudonym. (My identity isn't a secret—I’m just hesitant to make this my first Google hit, at least while I'm starting up!)
Why is this newsletter called “The Elbow”?
I’ll explain that in my second post. Smash that subscribe button ;)👇.