Research project Building a cross-cultural semantic framework for odor vocabularies
Describing smells is often difficult, yet research suggests that the vocabularies to express odor experiences vary widely between languages. Are odor vocabularies, despite this difference, similar on a deeper semantic level?

This project will investigate whether a shared odor-semantic space exists, akin to that of colors. Using a large-scale cross-linguistic study, we will analyze odor vocabularies of 25 culturally distinct languages.
The projects’ key objectives are 1) to characterize how the vocabularies differ in abstraction (the presence of dedicated odor terms) and codability (consistency in term use), 2) to map their semantic organization and identify common dimensions such as valence (pleasantness) and edibility, and 3) to develop a cross-cultural odor lexicon translatable across languages and representative of the perceptual odor space.
Through a collaboration in the Global Consortium for Chemosensory Research, we will collect data on odor descriptors and odor-perceptual ratings for 60 odors in 30 countries. We will also collect data on the descriptors’ semantic properties and odor applicability and systematically compare the languages' odor vocabularies.
Our findings will advance understanding of the interplay between language, culture, and odor perception and provide an empirically based resource for further research and practical applications. The project is interdisciplinary and offers novel insights into a fundamental yet underexplored aspect of human experience.
The sense of smell appears to be the hardest to verbalize in comparison to the other senses (Majid 2021; Olofsson & Gottfried 2015). Most western languages contain very few basic terms that are used to express olfactory experiences (i.e., they lack dedicated olfactory vocabularies, Majid 2021). At the same time, some “non-western” languages have well-developed olfactory vocabularies, and olfactory experiences are more frequently talked about in those languages (Majid 2021). Recent research suggests a major overlap in the semantic organization of the English smell expressions (e.g. Hörberg et al. 2022) and the dedicated olfactory vocabularies of non-western languages such as Maniq (Wnuk & Majid 2014), Jahai (Majid et al. 2018a) and Thai (Wnuk et al. 2020). The olfactory vocabularies of these languages are semantically organized in a similar way to English (e.g., Hörberg 2022), primarily differentiating between odors in terms of pleasantness, and secondarily in terms of edibility-“dangerousness”. Cross-linguistic differences observed in the olfactory domain might thus obscure a shared organization at a deeper level. Even though olfactory vocabularies show cross-linguistic differences at the structural and lexical levels, they might share an underlying universal structure at the semantic level. We hypothesize that languages might differ in terms of their preferred descriptive strategies for describing odors (e.g., in their preference of source-based vs. abstract terms), the specificity of their odor descriptors, and therefore, relatedly, their codability (see Majid et al. 2018b) of odors. However, the semantic organization of their odor vocabularies will be organized along the same dimensions, primarily distinguishing odors in terms of valence / pleasantness, and secondarily with respect to edibility / dangerousness or toxicity (Haddad et al. 2010).
In other to investigate this, we will use the Global Consortium for Chemosensory Research (GCCR) collaboration to collect cross-linguistic and cross-cultural data of olfactory descriptors and perceptual ratings of a set of selected odors. From this data set, we will derive language-specific olfactory lexicons with the most frequently used olfactory descriptors of each language. This data set will allow us to assess to what extent olfactory vocabularies differ across languages with respect to structural and lexical properties, such as their descriptive strategies (i.e., semantic category, see Hörberg et al. 2022), the specificity of their expressions, and their odor codability (as measured by e.g. Simpsons Index, see Majid et al. 2018b). On the basis of the GCCR collaboration, we will also collect cross-linguistic data of the semantic properties of the selected descriptors of each language. This data set will contain ratings of a number of semantic dimensions (e.g., valence, edibility, modality associations, concreteness) that have been found to be prominent in olfactory vocabularies in previous work (e.g., Hörberg et al. 2022), enabling us to determine to what degree the olfactory vocabularies converge and diverge on these dimensions.
With these two data sets, we will be able to determine whether odor vocabularies primarily differ in terms of superficial properties, but converge on their underlying semantic dimensions. For each odor × language combination, we will be able to compute the language- and culture-specific profiles of each lexical-semantic dimension. These profiles will together make up the multi-dimensional space of the olfactory vocabularies of each language. We will then be able to assess on which dimensions olfactory vocabularies converge, and on which they diverge. We hypothesize that language will diverge on "superficial" dimensions related to lexical and structural properties of the language, such as description strategy, specificity, simpsons index, but will converge on more "fundamental" dimensions related to semantic properties, such as valence, edibility and possibly concreteness. Importantly, the strength of these convergences will align with the "strength", "importance" or "salience" of these dimensions in the olfactory vocabularies of individual languages, on the one hand, and in the non-linguistic classification of odors, on the other. Olfactory vocabularies will thus primarily converge on the valence dimension and secondarily on the edibility dimension.
References
Haddad, R., Weiss, T., Khan, R., Nadler, B., Mandairon, N., Bensafi, M., Schneidman, E., & Sobel, N. (2010). Global Features of Neural Activity in the Olfactory System Form a Parallel Code That Predicts Olfactory Behavior and Perception. Journal of Neuroscience, 30(27), 9017–9026. 10.1523/JNEUROSCI.0398-10.2010.
Hörberg, T., Larsson, M., & Olofsson, J. K. (2022). The Semantic Organization of the English Odor Vocabulary. Cognitive Science, 46(11). 10.1111/cogs.13205.
Majid A. (2021). Human Olfaction at the Intersection of Language, Culture, and Biology. Trends in Cognitive Sciences, 25(2), 111–123.
Majid, A., Burenhult, N., Stensmyr, M., de Valk, J., & Hansson, B. S. (2018). Olfactory language and abstraction across cultures. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1752), 20170139. 10.1098/rstb.2017.0139
Majid, A., Roberts, S. G., Cilissen, L., Emmorey, K., Nicodemus, B., O’Grady, L., Woll, B., LeLan, B., de Sousa, H., Cansler, B. L., Shayan, S., de Vos, C., Senft, G., Enfield, N. J., Razak, R. A., Fedden, S., Tufvesson, S., Dingemanse, M., Ozturk, O., … Levinson, S. C. (2018). Differential coding of perception in the world’s languages. Proceedings of the National Academy of Sciences, 115(45), 11369–11376. 10.1073/pnas.1720419115
Olofsson J., & Gottfried J. (2015). The muted sense: Neurocognitive limitations of olfactory language. Trends in Cognitive Sciences, 19(6), 314–321.
Wnuk E., Laophairoj R., & Majid A. (2020). Smell terms are not rara: A semantic investigation of odor vocabulary in Thai. Linguistics, 58(4), 937–966.
Wnuk E., & Majid A. (2014). Revisiting the limits of language: The odor lexicon of Maniq. Cognition, 131(1), 125–138.
Kurfalı, M., Olofsson, J. K., Herman, P., Pierzchajlo, S., & Hörberg, T. (2025). Perception of smells; the next frontier for language models? Cognition, 264(2025), 106243.
Hörberg, T., Larsson, M., & Olofsson, J. K. (2022). The semantic organization of the English odor vocabulary using natural language data. Cognitive Science, 46(11), 1-31.
Study Protocol ManyLabs0002 pdf, 867.6 kB.
The project on other sites:
Global Consortium for Chemosensory Research