What is difficult to name in one language may be an ordinary distinction in another.
Look at something red.
Now name its color.
Red.
You do not need to say:
the color of blood.
You do not normally search your memory for a red object and then describe the color through that object.
English gives you a conventional abstract word for the perceptual quality itself.
Now smell coffee.
What does it smell like?
For many English speakers, the most natural answer is almost comically circular:
It smells like coffee.
Try another smell.
A lemon smells lemony.
Smoke smells smoky.
Fish may smell fishy.
A rose smells like a rose.
Again and again, instead of naming the olfactory quality itself, English points back toward a probable source.
That habit can feel so natural that we may mistake a property of English for a property of the human senses.
Perhaps smells simply cannot be named the way colors can.
Perhaps olfaction is inherently resistant to language.
Perhaps the nose gives us sensations for which words are naturally inadequate.
Then we encounter Maniq.
And the problem changes.
1. Maybe smell is not the problem
Maniq is an Aslian language spoken by a small hunter-gatherer population in southern Thailand.
For our purposes, one feature is particularly striking:
Maniq has more than a dozen terms dedicated to qualities of smell.
These are not merely names of odor sources.
That distinction is essential.
There is a difference between saying:
It smells like X
and possessing a conventional lexical item for a recurring quality of smell that can occur across different sources.
English certainly has some relatively general olfactory descriptions:
fragrant
musty
pungent
rancid
stinky
and others.
So it would be wrong to claim that English has no vocabulary for smells.
But English speakers very often identify unfamiliar odors through their sources or comparisons:
like garlic
like wet soil
like petrol
like rotten eggs
like smoke
The linguistic strategy frequently becomes:
odor → remembered source → description.
Maniq demonstrates that this is not the only way a language can organize olfactory experience.
2. Compare this with color
English speakers usually do not say:
This shirt is the color of grass.
We say:
green.
We do not normally say:
The car is the color of coal.
We say:
black.
Color vocabulary allows us to detach a perceptual property from a particular object.
A green leaf, green wall, green bottle and green light can all be described through the same abstract color category despite being completely different things.
That makes the category highly portable.
The question is therefore not simply:
Can humans perceive different smells?
Obviously they can.
The more interesting question is:
Can a language conventionalize abstract categories for those smells?
Maniq tells us that the answer is yes.
3. The smell word is not necessarily the smell source
This is the conceptual jump.
Suppose several completely different things share a recognizable odor quality.
A language can organize them in at least two broad ways.
It can repeatedly identify the sources:
X smells like A.
Y smells like B.
Z smells like C.
Or it can conventionalize a category that cuts across those sources:
A, B and C instantiate odor quality Q.
Once that happens, the smell itself becomes much easier to talk about independently of whatever happens to produce it.
This resembles what ordinary color vocabulary already does for English.
Red does not mean blood-colored.
Blood is one possible red thing.
The category has escaped its exemplar.
Olfactory vocabulary can do something similar.
4. Maniq smell words form a system
This is where the evidence becomes much stronger than an anecdotal list of unusual vocabulary.
Researchers Ewelina Wnuk and Asifa Majid investigated Maniq odor terminology experimentally.
They did not simply collect a few words and announce that Maniq speakers were “better at smelling.”
They examined how speakers understood relationships among the odor terms.
Three kinds of tasks were used, including listing exemplars, judging similarities and rating relevant properties.
The result was important:
the odor vocabulary behaved as a coherent semantic domain.
In other words, these were not miscellaneous descriptive expressions accidentally associated with smells.
There was internal semantic organization.
And two major dimensions emerged from the analysis:
pleasantness
and
dangerousness.
That second dimension is particularly revealing.
5. A smell can classify more than a sensation
From an English-speaking perspective, it is easy to imagine odor vocabulary primarily through a scale such as:
pleasant ↔ unpleasant
Perfume smells good.
Garbage smells bad.
But an olfactory system can participate in much more than aesthetic judgment.
Smell can provide information about food.
Plants.
Animals.
Health.
Environmental conditions.
Potential danger.
Suitable or unsuitable substances.
The Maniq evidence therefore forces us to stop treating smell as merely an impoverished sensory domain waiting for metaphorical assistance from other senses.
For the community using the language, olfactory distinctions can participate in a much broader network of practical and cultural knowledge.
The vocabulary does not merely answer:
Does this smell nice?
It can help organize:
What kind of smell is this?
and potentially:
What does this smell tell me about the world?
6. This does not mean Maniq speakers have better noses
Here we need the same caution that protected us in our discussion of Yélî Dnye.
A linguistic difference is not automatically a biological difference.
The existence of an elaborate odor lexicon does not by itself prove that Maniq speakers possess more sensitive olfactory receptors.
Nor does difficulty naming smells in English prove that English speakers literally smell less.
We have to separate several layers:
sensory detection
perceptual discrimination
conceptual categorization
lexical categorization
retrieval of a conventional word
cultural relevance
They interact.
They are not identical.
If two speakers can discriminate two odors equally well but only one language provides conventional abstract terms for them, the linguistic difference is real without requiring different noses.
7. Naming is not sensing
This distinction sounds obvious until we actually test it.
Imagine placing two unidentified substances in front of an English speaker.
The speaker smells both and says:
They definitely smell different.
Then ask:
What does the first one smell like?
A pause.
I don’t know… something woody? Maybe medicinal?
And the second?
Kind of sweet… but also… I don’t know.
The sensory distinction has already occurred.
The lexical identification has failed.
So:
I can distinguish X from Y
does not entail:
I possess stable lexical categories for X and Y.
And:
I cannot name X
does not entail:
I cannot perceive X.
That separation is fundamental to understanding sensory language.
8. English itself proves the point
Take color again.
Imagine someone shows you a color between blue and green.
You might hesitate:
turquoise? teal? cyan?
The uncertainty does not mean your eyes stopped working.
It means the perceptual continuum and the available lexical categories do not align perfectly at that moment.
Now imagine an entire sensory domain in which your language offers relatively fewer conventional abstract labels and repeatedly encourages comparison with sources.
The difficulty of talking about that domain could easily be mistaken for difficulty perceiving it.
But those are different problems.
9. “It smells like…” is already a linguistic strategy
This phrase deserves more attention than it normally receives.
When an English speaker says:
It smells like cinnamon,
the language is solving a naming problem through analogy.
The speaker identifies an experienced odor by locating a familiar source with a similar olfactory profile.
Conceptually:
current sensation → stored source representation → source name
This is an extremely effective communicative strategy.
There is nothing defective about it.
But it is a strategy.
And once we recognize it as a strategy rather than a natural property of olfaction, another possibility becomes visible:
current sensation → conventional odor category → odor word
Maniq shows that human language can travel much further along that second route than speakers of languages such as English might expect.
10. Why does “red” feel more objective than “smells like coffee”?
Because the lexical architecture is different.
When we say:
red,
the word appears to identify a property.
When we say:
smells like coffee,
the expression explicitly identifies a comparison.
This difference influences how the experience is packaged for communication.
One construction says:
THIS QUALITY.
The other often says:
THIS REMINDS ME OF THAT SOURCE.
Neither construction guarantees a more accurate perception.
But they place different linguistic tools between experience and communication.
11. This is the reverse side of the Yélî Dnye problem
In our discussion of Yélî Dnye, we encountered a crucial warning:
absence of a conventional word does not prove absence of a concept.
Maniq lets us approach the relationship from almost the opposite direction.
If a language develops a rich set of conventional words in a domain where another language relies heavily on circumlocution and source comparison, that does not prove that its speakers possess an entirely different sensory apparatus.
But it does show something important:
languages can differ radically in how much lexical structure they make readily available for talking about the same broad sensory domain.
Put the two cases together:
Yélî Dnye:
no familiar lexical boundary → do not infer missing thought.
Maniq:
rich lexical distinctions → do not infer a different human nose.
Between those two errors lies the interesting territory.
12. Language can make a distinction cheap
This may be one of the most useful ways to understand vocabulary.
Suppose a distinction can be expressed in both Language A and Language B.
In Language A, there is a conventional word.
In Language B, you need a phrase, comparison or explanation.
Both languages can communicate the idea.
But the communicative cost is different.
A conventional word makes a distinction:
shorter,
more reusable,
easier to retrieve,
easier to combine with other grammatical material,
and often more available as a recurring topic of conversation.
This does not imprison thought.
It changes the economics of expression.
That is a much more defensible claim than saying that language determines what humans are capable of perceiving.
13. Vocabulary is not a list of labels
This is where ordinary language teaching often undersells vocabulary.
A vocabulary list looks like this:
word — translation
word — translation
word — translation
But Maniq immediately creates a problem for that model.
If English habitually lacks an equally conventional category with the same semantic extension, what exactly should appear on the other side of the flash card?
A source?
A definition?
Several examples?
A descriptive phrase?
The difficulty is not accidental.
It reveals that lexical learning is not merely acquiring new labels for an already divided world.
Sometimes learning another language means discovering:
which distinctions have been made lexically reusable in that language.
That is a different task.
14. Translation can create a source where the original has a quality
Suppose an odor term in one language has no compact English equivalent.
A translator may have to render it through examples:
a smell like…
or through a descriptive phrase.
The English sentence can communicate successfully.
But something has changed.
The source language may have treated the odor as a conventional category.
English may reconstruct it through a comparison.
The referential result can be adequate while the semantic architecture changes.
This is similar to what we have repeatedly encountered elsewhere.
In Fijian possession, English my can flatten distinctions in the relationship between possessor and possessed.
In Plains Cree, an English translation can preserve who acts upon whom while making the direct–inverse architecture disappear.
Now smell gives us another possibility:
translation can preserve the sensory reference while replacing a lexical quality with a source-based description.
15. The dictionary question is therefore sometimes wrong
A learner asks:
What does this word mean in English?
The teacher searches for one English word.
Sometimes there is one.
Sometimes there isn’t.
When there isn’t, the temptation is to conclude that the foreign word is exotic, mystical or “untranslatable.”
But lack of one-word equivalence is much less dramatic.
It often means simply:
the two lexical systems package the semantic domain differently.
The correct teaching sequence then becomes:
What situations does the word describe?
Which examples belong to the category?
Which examples do not?
What semantic dimensions organize it?
How would English normally communicate the same distinction?
Only after that should we worry about a compact translation.
16. Smell exposes a bias in how we imagine language
Vision dominates many everyday metaphors for knowledge.
We say:
I see.
That’s clear.
Look at the problem.
What’s your point of view?
Even our descriptions of understanding repeatedly recruit visual language.
That can make the vocabulary of vision feel like the natural model for perceptual language.
Color words seem ordinary.
Shape words seem ordinary.
Spatial descriptions seem ordinary.
Smell words can seem vague.
Maniq reminds us that this hierarchy should not automatically be projected onto human language as a universal.
A sensory domain that one language treats awkwardly can be linguistically elaborated elsewhere.
17. Does culture matter?
Yes — but this is another place where explanation can become too easy.
It would be tempting to say:
The Maniq environment makes smell important, therefore the language has smell words.
That may capture part of the story, but it should not be turned into a simplistic causal law.
Languages, practices, environments and communicative needs interact historically.
We should not imagine that a community collectively examined its environment and deliberately designed the necessary vocabulary.
Nor should we assume that every culture in a similar ecology must develop the same lexical system.
The safer and more interesting observation is that Maniq olfactory terminology participates in a broader cultural ecology in which smell has considerable practical and cultural relevance.
Language is part of that ecology, not a mechanical reflection of it.
18. A smell word can carry knowledge
Once an odor category becomes conventional, it can accumulate associations.
Not merely:
What does it smell like?
but:
Where does this smell occur?
What produces it?
Is it desirable?
Is it dangerous?
What does its presence predict?
This is where vocabulary becomes more than naming.
A lexical category can become an access point into accumulated knowledge about recurring patterns in the environment.
The word is short.
The network behind it may be enormous.
19. What exactly does language change?
We can now answer more carefully.
Language does not need to manufacture a sensory experience in order to matter.
It can influence how readily an experience is:
categorized,
named,
remembered linguistically,
compared,
discussed,
taught,
and integrated into shared knowledge.
Those are substantial effects.
We do not need the stronger and much harder claim that language determines perception itself.
The interesting linguistic phenomenon is already sitting in front of us.
20. From “What does it smell like?” to “What smell is it?”
That tiny change of question captures almost the entire article.
English very naturally asks:
What does it smell like?
Notice the structure.
The question itself invites a comparison.
Something smells like something else.
Now imagine a linguistic environment in which an odor can more readily be treated as a category in its own right.
The communicative problem changes.
We move from:
Which familiar object does this sensation resemble?
toward:
Which olfactory quality is this?
The nose may be encountering the same physical world.
The language is giving the speaker a different route through it.
21. The lesson is not that Maniq is richer than English
Languages cannot sensibly be ranked by counting lexical distinctions in isolated domains.
English has enormous terminological resources in domains where other languages may organize things differently.
Maniq elaborates olfaction in ways particularly useful for understanding how sensory vocabulary can work.
Neither fact makes one language superior.
The point is comparative.
Every language makes some distinctions easy to package.
Other distinctions require more work.
And because we spend most of our lives inside those habitual solutions, we can easily mistake them for the only possible architecture of meaning.
22. Thinking instead of memorizing
For a language learner, this changes the question completely.
Do not begin with:
What English word equals this word?
Begin with:
What distinction is this language making?
Then ask:
What experiences belong together under this word?
Where are the boundaries of the category?
Would my language use one word, several words, a comparison or an explanation?
What disappears if I replace the category with the nearest translation?
That is vocabulary learning at the level of meaning rather than substitution.
And once you start asking those questions, even something as ordinary as a smell becomes evidence about how language works.
Conclusion: the nose was never speechless
It is easy for speakers of English and related languages to experience the difficulty of naming odors and conclude that smell itself resists language.
Maniq shows why that conclusion is too large.
Human beings can conventionalize recurring olfactory qualities.
Those qualities can form a structured semantic domain.
They can be related to pleasantness, danger and broader systems of practical and cultural knowledge.
None of this proves that Maniq speakers possess fundamentally different noses.
And the relative difficulty English speakers experience in naming odors does not prove that their noses perceive less.
The difference lies somewhere more interesting:
between sensation and the linguistic categories available for making that sensation communicable.
Yélî Dnye showed us that a missing word does not necessarily mean a missing concept.
Maniq gives us the complementary lesson:
when another language names distinctions that yours usually leaves unnamed, the world has not suddenly acquired new sensations. You have discovered another way of organizing them.
The limits of your vocabulary are not automatically the limits of perception — but another language can reveal distinctions your vocabulary taught you not to name.
Tymur Levitin
Founder & Director, Levitin Language School
Certified Translator and Language Teacher
© Tymur Levitin / Levitin Language School. All rights reserved.
