In Cantonese, the syllable “si” can mean a poem, a page of history, an attempt, a season of time, a market, or the verb to be. The consonant and the vowel stay put. What changes is only the pitch of the voice, whether it rises, falls, dips or holds flat, and that alone is enough to swap one word for another.
This is not a quirk of a single language. Cantonese carries six tones by the most common count, and the habit of leaning on pitch to separate meanings is shared by a large share of the world’s languages, from Mandarin and Vietnamese to Yoruba and Zulu. Linguists group these as tonal languages, and those that contrast more than two tones are classed as complex tone.
What is less widely known is where these languages sit on a map.
Where tonal languages cluster
Complex tone is spread unevenly across the planet. It concentrates in warm, humid regions closer to the equator: sub-Saharan Africa, mainland Southeast Asia, southern China, and pockets of the Amazon basin and New Guinea. It is comparatively scarce in cold or arid zones, across northern Europe, Central Asia, the drier parts of the Middle East, and the Indigenous languages of interior Australia.
A distribution this uneven invites a question about cause.
The humidity hypothesis
The most cited attempt to explain it comes from a 2015 paper in the Proceedings of the National Academy of Sciences, “Climate, vocal folds, and tonal languages: Connecting the physiological and geographic dots,” by Caleb Everett of the University of Miami with Damián Blasi and Seán Roberts, then at Max Planck institutes in Germany and the Netherlands. The authors drew on laryngology research indicating that dry air affects the vocal folds, reducing the elasticity and fine control needed to shift and hold pitch cleanly. Dry conditions, the paper reports, are associated with increased jitter and shimmer, small irregularities in how the vocal folds vibrate.
From that physiology they made a prediction. Languages that depend on precise pitch should be less likely to develop in very dry or very cold places, and more likely to develop where the air is humid.
What the study actually looked at
To test the idea, the authors compared the geographic distribution of tonal and non-tonal languages against humidity data, using two large, independently compiled databases covering more than 3,700 languages. Complex tone, they found, was largely absent from the coldest and driest regions and clustered in humid ones, in line with what the physiological argument predicted.
Everett was careful about the size of the claim. Speaking to Phys.org when the paper appeared, he said the work was not arguing that sound systems evolve entirely, or even mainly, because of climate, and that many cultural and language-internal factors shape how languages change over time. On this account, humidity acts as a gentle pressure across long stretches of history.
What the correlation can and cannot show
This is one line of research within an active debate. The correlation is drawn from a large sample, but any link between climate and language distribution is hard to separate from history: related languages tend to be spoken in neighbouring places, and those places tend to share a climate. So a climatic signal can partly reflect ancestry. They applied statistical controls to address this, and linguists have since argued over how much of the pattern survives them.
The finding is worth taking seriously, but it should not be read as the final word. It is a hypothesis with supporting data, presented by its own authors as a tendency rather than a rule.
What it leaves open
The distribution itself is easy to state: tonal languages are more common in the warm, wet parts of the world and rarer in the cold and dry ones. Whether humidity helps explain that shape over many generations, or whether it comes down mainly to migration, language contact and shared ancestry, is the part still being worked out.
