Language and Ecology — Distribution of Tonal Languages

Previously, we looked at the ecological drivers behind global patterns and distribution of language diversity. On a global scale, language diversity increases towards the equator, and the tropics are found to be more language dense. To explain these patterns from an ecological perspective, it was proposed that ecological risk played a more significant role than landscapes or isolation in influencing these patterns. While these two factors cannot fully explain the patterns observed, the authors proposed sampling biases which could explain the seemingly language-poor Amazon after accounting for the ecological drivers. Higher in situ diversification rates was also proposed to explain the seemingly language-rich regions. The lack of detailed studies surrounding distance-based and dated phylogenies on a global scale makes it difficult to properly assess this proposition, however.

When we look closer in global patterns of language diversity, we spot another trend. Tonal languages, languages where words can differ in phonetic tones to give different meanings, are found to be highly concentrated in tropical regions, while generally absent in arid climates like deserts. Tonal languages, particularly those with complex tones, are rare and accounting for less than 10% of the world’s languages, according to the Phonotactics Database of the Australian National University. Nevertheless, the global distribution of these languages produce an interesting pattern, with seemingly obvious associations. Tonal languages are more likely to be found in hot, humid regions like Southeast Asia, New Guinea and Sub-Saharan Africa, and highly unlikely to be found in arid regions like the Gobi, Australian, Saharan and Arabian deserts. Could air humidity possibly have played a role in shaping the distribution of tonal languages, and what implications does this have on environmental drivers of human phonetics and phonology? Here, we introduce a study done by Everett, Blasi and Roberts (2015), who aimed to assess the correlation between environment and phonetics in tonal languages.

Before this study, it was thought that human sound systems were not really ecologically adaptive, meaning that the environment did not have a significant impact on influencing the sounds we produce. Even when correlations or patterns were revealed, the hypotheses proposed for some mechanisms were not supported by experimental evidence, leaving some to speculate that these patterns could be just epiphenomenal. Experimental research on the human larynx has been ongoing for quite some time by the time this paper was published, and it was from these experiments that aided the prediction of correlation between environment and human phonetic tone production.

Phoneticians found out that the prevalence of tones were more associated with linguistic areas than families, reflecting a more significant geographical correlation than phylogenetic or genealogical. The team used mean specific humidity and mean annual temperature to map out the correlation patterns, leading to the finding. Languages with complex tone are least likely to be found in areas of low mean specific humidity, and more likely to occur in areas of high humidity. Interestingly, they also found that languages with complex tone are less likely to occur in cold regions with subzero mean annual temperatures. These patterns are generally not influenced by genealogical relationships between languages, after factoring in the existence of language isolates like Ticuna, Laal and Morori. The authors suggested that tones spread across languages through cross-language contact in more favourable environments (humid, hot), and that cold or arid regions functioned as a barrier to this spread of complex tone. This raises the question: why does this happen?

Tonal languages normally necessitate voicing at pitches throughout a person’s normal vocal range, like 100Hz for a typical adult human male. Lexical tones can differ by 10Hz, and phonemic contrast in tones, like falling or rising pitch can differ by 20-30Hz. We see languages with high, mid and low tones like Thai, and the high, low, creaky and checked tones in Burmese. Creaky and checked tones are regarded as falling tones, but they are distinguished by other factors like length, intensity and breathiness, which can involve altering the laryngeal settings to produce them. This calls for the biomechanics of speech to be taken into account.

The larynx, also known as the voice box, has vocal folds and their physiology is directly influenced by hydration levels, particularly in the viscosity and elasticity of vocal folds. Through previous laboratory studies of the larynx, laryngologists (people specialised in studying the voice box, and associated disorders and diseases) found that under increased hydration, as one would normally see in humid regions, there is an increased vocal fold viscosity and ability to facilitate phonation, while under desiccation, they recorded increased perceived phonation effort in the larynx, affecting jitter and shimmer, terms to reflect imprecise pitch and varying amplitudes respectively.

Let’s say a person inhales dry air, or air strongly lacking in moisture. This affects the vocal fold physiology, decreasing the viscosity of mucus in the larynx and affecting various cohesive forces used to produce vocal fold oscillations. Even in healthy individuals, breathing in dry air could impair voicing abilities slightly. Dry air was also found to be the cause of sore throats in a substantial number of cases, reflecting the impact dry air has on the larynx. In fact, singers also tend to be less able to maintain tones or pitches in dry conditions, reflecting the impact of aridity on the ability to produce and maintain tone, let alone communicable tone.

How does this translate to this study? We noted that subzero temperatures and dry regions are areas least likely to support languages with complex tone. Frigid air is less able to hold water vapour than hot air, so even if frigid air of say, -10 degrees, is at maximum saturation, it holds considerably less water vapour than air at 30 degrees. This implies that cold air, especially in subzero temperatures, is almost always dry.

Noting that dry air has deleterious consequences on the ability to produce precise phonation for tones, we can start to piece this together with geographical patterns, possibly explaining to some extent the distribution of languages with complex tone. Some might question if there is any genetic consequences from the effects of air quality on the human larynx. It turns out, not much. Studies on sheep and dog larynges show similar patterns in voicing phonations, suggesting that this phenomenon is genotypically and phenotypically normal across all human populations.

However, we have to make a special distinction. Speakers of languages with complex tone can still communicate with each other in frigid areas like Antarctica. What this study suggested is that languages are far less likely to develop complex tones if they are in frigid or dry regions. This however, does not entail that languages should and would develop complex tone in warm and humid regions over time.

This study could prompt the study of evolution of languages. Reconstruction of languages like Mandarin Chinese suggested that Old Chinese, spoken from the Shang dynasty up until the Qin unification, lacked any form of tones. Questions remain over the plausibility of this reconstruction theory, but if this is correct, it might reveal a bit about the climate and environment then to support the development of complex tones in the Mandarin Chinese we speak today. The issue, though, is that this study did not make any indication of the time scale involved in this adaptation. This study, just like the one discussed previously, captured a snapshot of the distribution of the world’s languages given the current climatic data. Perhaps 1000 years ago, this scene could be way different, with a slightly different climate, but it is difficult to extrapolate to that period. Nonetheless, this study still offers meaningful thought on the role of the environment in the sounds we produce, prompting the study of ecological adaptations in other human sounds.


Here we see how the development of language features like complex tones over time can be influenced by the environment, which is a rather interesting pattern to look at. Over time we might uncover more patterns surrounding distributions of the world’s languages, and perhaps develop more robust explanations to explain the patterns.

Further Reading

Everett, C., Blasi, D. E. & Roberts, S. G. (2015) Climate, vocal folds, and tonal languages: Connecting the physiological and geographic dots. Proceedings of the National Academy of Sciences, 112 (5), 1322-1327. Available from:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s