Research indicates that people worldwide link the trilled R sound with a rough texture and a jagged shape, while the L sound is associated with a smoother texture and a flatter shape. Researchers propose that this relationship may be even more widespread than the well-known bouba/kiki phenomenon.
A new study from the University of Birmingham (UK), published on November 20th in the Journal of the Acoustical Society of America, presents one of the strongest examples of “sound symbolism”– the direct relationship between speech sounds and their meanings.
Marcus Perlman, an Associate Professor specializing in Linguistics and Communication at the University of Birmingham, stated: “Our research indicates that speech sounds produce a sense of texture and shape. In an earlier study, we discovered that R sounds are prevalent in adjectives describing rough surfaces rather than smooth ones across various languages. This phenomenon may stem from a type of similarity between the sound of a word and the texture it describes. In this investigation, our aim was to explore the perceptual link between the trilled R sound and the notion of roughness.
“The bouba/kiki effect is the most recognized instance of sound symbolism in language, where made-up words like ‘bouba’ match with round shapes, while ‘kiki’ aligns with angular shapes. We previously showcased the bouba/kiki effect through experiments with speakers of many languages, but the R/L effect demonstrates an even stronger and more consistent association across different cultures. These types of cross-modal correspondences could have played a key role in the development of spoken languages, influencing the terminology we use to describe textures and shapes.”
From Albanian to Zulu
The researchers carried out their study using both online and field experiments with 1,030 adults fluent in 28 languages, including Zulu, Albanian, Danish, English, Greek, Italian, Farsi, Spanish, Russian, Japanese, Korean, Mandarin Chinese, Thai, Daakie, and Palikúr. Participants viewed images of two lines—one jagged and one straight—and visualized running their fingers along each line. They then listened to a recording featuring a trilled R (like the rolled “R” sound in Spanish) and an L sound, matching each sound with one of the lines.
The findings revealed that a significant majority of participants consistently matched R with the jagged line and L with the straight line, regardless of the order in which they were presented. The strongest correlation was found between the trilled R and the jagged line, with 94% matching on independent trials, followed by 84% for matching L to the straight line. Speakers of languages such as Estonian and Finnish exhibited a perfect match rate of 100%, while speakers of languages including Albanian and Mandarin Chinese had the lowest rate at 70%.
The association between the trilled R sound and roughness was evident across speakers of all languages, even if that sound was absent from their language. For instance, Palikúr speakers, whose language doesn’t include a trilled R, still achieved a 100% matching rate. Additionally, even speakers of languages like Mandarin Chinese and Japanese, which do not frequently use the trilled sound, perceived it as rough and jagged.
A Nearly Universal Connection Between Speech and Texture
The study presents one of the most compelling instances of iconic relationships between speech sounds and sensory properties related to touch and vision.
Bodo Winter, a Professor of Linguistics at the University of Birmingham, remarked: “The results from our experiments are even more consistent than those from the cross-cultural studies of the bouba/kiki effect. Almost all participants in our study identified R as rougher than L, showing a 15% stronger association compared to the bouba/kiki effect. This pattern was consistent across cultures, with every language group displaying the same link. This differs from bouba/kiki, which does not hold true in some languages. Thus, the connection between R/L and roughness/smoothness may represent one of the most robust examples of sound symbolism documented to date.”
Moreover, while the overall matching rate across languages was high, participants whose primary language employs the alveolar trill R showed a slightly lower correct matching rate (86.6%) compared to those whose native language does not use this sound as the main variant (89.8%). This suggests that the traditional usage of the sound could reduce its iconic impact.
Dr. Perlman concluded: “The trilled R is one of the most intriguing speech sounds. It’s notoriously challenging for speakers to master, and some languages even have specific terms for individuals who cannot pronounce it (like “Erre moscia” in Italian). Yet, trilled Rs are surprisingly widespread in various languages. It’s puzzling why such a difficult sound is so common. We believe the iconic nature of R contributes to this. The sound conveys a strong sense of texture and shape, possessing a remarkable expressive quality that encourages its use, despite the difficulty of articulation. Additionally, it’s an exciting sound and fun to produce!”
