Sometimes, a word can seem absolutely perfect for the concept it describes. How else would you describe leaves moving other than “rustling”? What better captures a grim mood than “sulk”? Or, to paraphrase Shakespeare, would a “rose” by any other name smell as sweet? Humans naturally connect language with the real world — language is our main tool for understanding and organizing the objects around this — but it’s up for debate how much a word’s sound and meaning are really connected.
For a long time, the accepted notion among linguists is that words are arbitrary. That means there’s no reason why we call a fish a “fish” or a wrench a “wrench”; they’re all just random choices made by humans hundreds of years ago. And yet, this notion has been losing ground in recent years.
Recent studies — as well as some older ones — have shed some doubt on the idea of language being entirely arbitrary. The truth is still murky, but looking at the topic through the lens of the brain can provide a fascinating look at how language works.
The most common counterexample to the arbitrary nature of words is onomatopoeia, or words that specifically mimic a sound in the real world. A dog “barks” and a bee “buzzes” and a doorbell goes “ding-dong,” and all of these words were chosen because they resemble the actual sound that these animals and objects make. In that case, they’re not arbitrary. Instead, they’re an example of iconicity, which refers to the idea that certain words are indeed connected to the real-world concept they describe. In recent years, iconicity has been gaining ground.
For some linguists who still argue that words are indeed arbitrary, onomatopoeia is an exception, rather than an entire disruption of the initial linguistic notion. In that case, yes, onomatopoeia is not entirely arbitrary, but onomatopoeia is a very small subset of the total number of words in any given language.
For others, onomatopoeia doesn’t present any problem to the initial idea at all: their argument is that onomatopoeia is also arbitrary. That may seem counterintuitive. A duck saying “quack” makes sense, and a duck saying “fabulous” doesn’t. When you compare onomatopoeia across languages, however, there are some vast differences. Dogs in Spanish, for example, say guau-guau, whereas no English word for dog sounds has a “g” in it at all. If onomatopoeia weren’t arbitrary, wouldn’t that mean that these words would be consistent across all human languages?
It’s difficult to empirically prove the connection of sound and meaning using onomatopoeia. While it provides an interesting case study, it’s only a small part of the larger issue. To address that, you have to look at other patterns in language.
In 1930, linguist J.R. Firth came up with a new concept: the phonestheme. This is when a certain word sound is used in a number of words that are related. The most common example of this is the sound gl-, which appears in a number of different light-related words, including glitter, glisten, glow, gleam, glare and so on. The argument here is that something in the sound gl- is associated with light in the human brain, which would mean there’s something going on that isn’t arbitrary.
We’ve already looked at iconicity, but phonesthemes might suggest something else: systematicity. While it’s another academic-sounding word, at its simplest it just means that humans create sound-based systems that connect to a word’s meaning. So even though the gl sound doesn’t have a real-world connection to the concept of light, we people have created a connection between the two. If this is true, then you can’t say that those gl- words are entirely arbitrary. Systematicity and iconicity can work side by side, and they both challenge the notion of arbitrariness.
Recent studies are indeed finding examples of systematicity in language. The idea that humans use systems of sound to organize meaning, while still controversial, is less so than iconicity. Humans make patterns from everything, so it’s not too surprising that sounds could be associated with symbols in our brains. While this shows another crack in the “fully arbitrary” theory of word creation, it might just be another small exception to the arbitrary rule.
Bouba And Kiki
You can’t talk about the connection of sound and meaning without mentioning one of the most famous linguistic phenomena of all time. First explored by psychologist Wolfgang Köhler in 1929, the bouba/kiki effect caught more attention when it was brought into the limelight by a 2001 study by Vilayanur S. Ramachandran and Edward Hubbard. Part of its popularity is its simplicity. Study participants were shown two abstract objects — one with spiky edges and one with rounded edges — and then told that in a Martian language, one is called “bouba” and one is “kiki.” They’re then asked which one is which, and 95 percent associated “bouba” with the rounded figure and “kiki” with the spiky one.
This study in particular helped launch the re-examination of arbitrariness in language. After all, the bouba/kiki effect suggests that there might be something innate in humans that associates certain sounds with real-world concepts. Do we think of “kiki” as being spiky because the way we pronounce “k” is sharp? Is “bouba” the rounded shape because the letters of “bouba” themselves are rounded? Does this reveal some sort of proto-language underlying every human language in the world?
The bouba/kiki effect has shown enormous promise in reshaping how linguists think about sound and meaning. The effect was reproduced across at least 17 different languages, meaning it’s not some quirk of English speakers. It’s important to include one caveat, however, which is that it hasn’t been observed in every language. Romanian, Mandarin Chinese and Turkish speakers all assigned “bouba” and “kiki” to the non-expected shapes more than 50 percent of the time. While this is only three of the 25 languages tested in one study, it’s enough to suggest that this sound association isn’t universal.
While bouba/kiki has become shorthand for the phenomenon, we should also point out that it extends beyond these two nonsense words. One other common example is the trilled R sound — all but vanished in English, but common in other languages — which is associated with roughness. This is another data point suggesting that the way humans pronounce sounds (trilled R “feels” rough) could influence the words they appear in. Linguists are still figuring out exactly how prevalent the bouba/kiki effect is, however, and there are many paths that studies will explore in the future.
The Future Of Sound And Meaning
There have been a lot of advancements in the research of sound and meaning in the last few decades. There are some pretty strong indications that systematicity and iconicity are chipping away at the idea that all words are arbitrary. The main question remaining is just how much human perception affects words and how much words affect human perception.
Part of the reason that disentangling sound and meaning is so difficult is because humans are surrounded by language at birth. It’s very hard to figure out correlation and causation in language. Take, for example, color-grapheme synesthesia, which is when people see letters as colors (A is red, B is blue and so on). A study found that about 6 percent of people with this kind of synesthesia developed it from a specific model of Fisher-Price alphabet magnets. Without consciously knowing it, these people were absorbing language associations from a children’s toy.
You can imagine how hard it would be to figure out if words were chosen because humans associated them with real-world concepts, or if the words were chosen arbitrarily and then humans subconsciously created the connection. The best way to do this, really, is to continue doing studies that use participants who were raised in different places speaking different languages. Some of the most misleading results in this research come from a study that takes too narrow a scope, making it more susceptible to overlooking specific cultural effects.
As it stands now, it seems that arbitrariness is still the dominant factor in the development of languages. It would take quite a lot of evidence and research to overturn this notion. Even if that never happens, research into the topic will continue revealing interesting patterns that could give us insight into how languages formed in the first place.