Studies of SE Asian tribes force a re-think on the psychology of language and smell

“What does coffee smell like?” “What about lemon?” These questions are tricky for English speakers to answer because we tend to describe smells by referring to their typical source. So, an aroma that smells like coffee is described as, well, smelling like coffee. Ditto for lemon or cinnamon or rotten eggs.

The fact is we don’t have abstract words to describe the essence of these odorous experiences. This contrasts with our language for other sensory experiences such as colour. For example, the word “red” describes the first-hand experience of seeing that particular wavelength of light, and the word can be meaningfully applied regardless of the source of the red.

“What does this smell of?” is another difficult question for English speakers. When we’re asked to identify everyday items like coffee or chocolate by their smell, our accuracy is around 50 per cent. “Similar performance with a visual object … would [lead a person to] be diagnosed as aphasic and sent for medical help,” say Asifa Majid and Niclas Burenhult, the authors of a new investigation into the links between language and smell.

These researchers explain how the lack of smell-based terms in English, combined with our poor skills at smell identification, have led generations of scholars to propose that olfaction (the scientific name for the sense of smell) is unimportant to humans. Even great minds like Darwin and Kant have arrived at this conclusion.

But now Majid and Burenhult have uncovered evidence from a Malaysian tribe – the Jahai who speak Jahai – that shows such conclusions are premature and an over-generalisation. In a smell identification test, 10 members of the Jahai (all men; average age 37) were as precise naming 12 smells as they were at naming colours. In contrast, 10 age-matched speakers of American English were vague and inconsistent at naming 12 smells, but excelled as you’d expect at naming colours. What’s more, the Jahai were far more succinct than English speakers at naming the smells, and 99 per cent of the time they used abstract terms for smells (whereas this was rare for the English speakers).

The lesson from this field trip is that we shouldn’t assume that findings about language and smell from the Western world (especially English speaking cultures) necessarily apply to humanity as a whole. Odour is incredibly important to members of the Jahai, and they use at least 12 abstract smell-based words to discuss plants and animals in their everyday lives. “Jahai speakers show us that olfactory abstraction is possible,” said Majid and Burenhult, “and humans can be adept at talking about smells.”

What is it like to have a rich language for smells? Further insight comes from a study of another South East Asian Tribe – the Maniq in Thailand, who speak Maniq. The results of this investigation have been shown exclusively to the Research Digest ahead of publication. In the course of three field trips working with co-author Ewelina Wnuk, Asifa Majid first asked 8 Maniq speakers (4 female) to provide examples of items that fit 15 smell terms used in their language. This showed how the same term could be used to describe the same odorous essence attributed to a variety of different types of object, location or activity. For instance the term caŋə was attributed to food, cooked meat, and white sun, while paʔ ʔ was applied to old shelters, mushrooms and pouring water (among other things).

Next Wnuk and Majid tried to identify the factors underlying this diverse smell-based vocabulary. Members of the Maniq were presented with three of their smell terms at a time and asked to pick the odd one out. Successive trials of this kind suggested that the range of smell terms are best described as existing along two dimensions – pleasantness and dangerousness. Another study found strong consensus between tribe members about the extent to which each odorous term conveyed dangerousness, pleasantness and other descriptors.

As with the Jahai, the richness of smell-based language in Maniq reflects the lives of this tribe. They regularly discuss smells; they avoid bad – dangerous or unpleasant smells – and they deliberately use and wear appealing odours to bring good health and avoid danger. The attribution of smell terms to their cultural practices is often complex. For instance the term Caŋus, associated with pleasantness and cleanliness is also applied to the smell of the fruit “kul w”, even though it is poisonous. A leaf monkey – a food source for the Maniq – is said to smell of Caŋus if it has eaten the poisonous fruit – a warning that eating the monkey will cause sickness.

Research with the Maniq shows again that assumptions about the limited role of smell in our lives, and about our poor ability to describe smell experiences, has likely been premature. By studying exclusively Western samples we underestimate the richness of human experience. “The cultural and linguistic elaboration of smell among the Maniq constitutes compelling evidence against the universal paucity of olfactory terms, the ‘weak link’ between smell and language, and the general insignificance of olfaction for humans,” write Wnuk and Majid.


Majid A, and Burenhult N (2014). Odors are expressible in language, as long as you speak the right language. Cognition

Ewelina Wnuk, and Asifa Majid (In Press). Revisiting the limits of language: The odor lexicon of Maniq. Cognition.

further reading
The smell of fear more powerful than previously realised
Humans don’t smell that bad
Humans can track scents like a dog
Mice and humans like the same smells

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.