Describe a banana. It's yellow, perhaps with some green edges. When peeled, it has a smooth, soft, mushy texture. It tastes sweet, maybe a little creamy.

And it smells like... well, it smells like a banana.  

Every sense has its own “lexical field,” a vast palette of dedicated descriptive words for colors, sounds, tastes, and textures. But smell? In English, there are only three dedicated smell words—stinky, fragrant, and musty—and the first two are more about the smeller's subjective experience than about the smelly thing itself.

All of our other scent descriptors are really descriptions of sources: We say that things smell like cinnamon, or roses, or teen spirit, or napalm in the morning. The other senses don't need these linguistic workarounds. We don't need to say that a banana “looks like lemon;” we can just say that it's yellow. Experts who work in perfume or wine-tasting industries may use more metaphorical terms like decadent or unctuous, but good luck explaining them to a non-expert who's not familiar with the jargon.

Some scientists have taken this as evidence that humans have relegated smell to the sensory sidelines, while vision has taken center-field. It's a B-list sense, deemed by Darwin to be “of extremely slight service.” Others have suggested that smells are inherently indescribable, and that “olfactory abstraction is impossible.” Kant wrote that “Smell does not allow itself to be described, but only compared through similarity with another sense.” Indeed, when Jean-Baptiste Grenouille, the protagonist of Perfume: The Story of a Murderer can unerringly identify smells, remember them, and mix and match them in his head, he seems disconcerting and supernatural to us, precisely because we suck so badly at those tasks.

But not all of us. In Southeast Asia, there are at least two groups of hunter-gatherers who would turn their noses up at this textbook view. Asifa Majid from Radboud University in the Netherlands has found that the Jahai people of Malaysia and the Maniq of Thailand use between 12 and 15 dedicated smell words.

“These terms are really very salient to them,” she says. “They turn up all the time. Young children know them. They're basic vocabulary. They're not used for taste, or general ideas of edibility. They're really dedicated to smell.”

For example, ltpit describes the smell of a binturong or bearcat—a two-meter-long animal that looks like a shaggy, black-furred otter, and that famously smells of popcorn. But ltpit doesn't mean popcorn—it's not a source-based term. The same word is also used for soap, flowers, and the intense-smelling durian fruit, referring to some fragrant quality that Western noses can’t parse.

Another word is used for the smell of petrol, smoke, bat droppings, some species of millipede, the root of wild ginger, the wood of wild mango, and more. One seems specific to roasted foods. And one refers to things like squirrel blood, rodents, crushed head lice, and other “bloody smells that attract tigers.”

These terms don't refer to general qualities that are the dominion of other senses, like edibility. “Their meaning is not general over tastes, textures, pain, or any other state; their business is smell,” says Majid. “Just as you would describe a tomato as red, a Jahai speaker would describe the smell of bearcat as ltpit.”

Majid first heard about the Jahai through her colleague Niclas Burenhult, who had been working with them for years and had written the only formal grammar of their language. He noticed that they had specific smell words and, in disbelief, she flew out to Malaysia to investigate.

She talked to them, got to know their language, and learned about their culture. She tried them on the Brief Smell Identification Test—essentially a standardized scratch-and-sniff test with some fixed smells. She went on jungle walks with them, and asked them to describe the smells of their environment. And through several tests, she showed that they can name smells as consistently, easily, and clearly as English speakers can name colors, with none of the verbal struggles or linguistic somersaults that Westerners go through. The same applies to the Maniq.

These two groups clearly show that odors, contrary to popular belief, are not universally ineffable. “This work has changed my views!” says Tim Jacob from Cardiff University. “Smell information translates straight into behavior or mood and evokes whole memories,” he says. “Smell doesn't need language. That is what I thought. But, clearly there are cultures who have developed a smell vocabulary and it is both useful and necessary.”

Or, as Majid wrote, “Odors are expressible in language, as long as you speak the right language.”

And if you have the right language, it changes the way you perceive the world. Smell is an intrinsic part of Jahai culture in a way that it simply isn't in the west. “When we're there, Niclas and I say that we're brother and sister. Once, we got told off for sitting too close to each other because our smells would mingle, and that's a form of incest,” she says. “There are social taboos that are explained in terms of smells. Some foods can't be cooked in the same fire because their smells would mix.”

They're also extraordinarily good at distinguishing smells, and certainly better than Majid was. “We did a jungle walk during wild ginger flowering season,” she recalls. “You'd think it's one object and one smell but it's not. the flower smells different to the stem, and the stem smells different when you press it than when it's broken. I was trying to name the smells and some kids were following me around and laughing. Like: How can you be such a moron?

She is now planning to return to the tribe next year to work out whether they're better at remembering smells. She's also thinking about collecting odor samples of things that they describe with the same words, to see if she can isolate chemicals that are common to them all. And she wants to compare these languages, and those from other Southeast Asian tribes, to understand how these smell words evolved from earlier parent languages.

“This work is very important for going beyond the WEIRD subjects,” say Charles Spence from the University of Oxford. By that, he means people from western, educated, industrial, rich, and democratic countries—the ones who make up almost all of psychological research. “Anthropologists have been saying that we should do this for years but psychologists mostly ignore this advice. To have these beautiful examples where smell really is elevated from the bottom of the hierarchy is great.”

Jacob also thinks that the Jahai and Maniq could provide clues about the evolution of our olfactory sense. For example, it's obvious why we might be repelled by disgusting smells—they're associated with decaying food, feces, and other things that might carry disease. But why do we like pleasant smells? The Maniq provide a clue: Their bearcat word is also used for medicinal plants with pleasant smells, plants that they wear as perfumes and necklaces.

“This is, to me at least, revolutionary,” says Jacob. “Early cultures collected pleasant-smelling plants because they were medicinal, curative, or represented cleanliness and hygiene. Here, at last, is a positive function of pleasant smell that confers adaptive advantage. What's more, it points to the origins of our fascination with perfume.”

Indeed, as Majid says, “We spend millions on flavor and scent industries. To think that smell isn't relevant at all for humans is too simplistic.”