There are a bunch of tonal languages spoken throughout the world, but by a large margin, Mandarin Chinese comes up at the top in terms of number of native speakers. But its size or scope of usage is not the focus of our discussion today. One thing that intrigues me is the history of the Chinese language, from the evolution of the writing system, to the deduction or theorisation of what Mandarin Chinese might have sounded like centuries back.

Previously, we have covered a bit on the Qieyun, a rhyming dictionary that allowed us to look at how Mandarin might have sounded like in the past. However, another aspect that I have wondered about, even as a native Mandarin Chinese speaker, is how we got four tones. Looking at some other related languages, Hakka has six tones, Hokkien has five or six tones depending on the dialect, Suzhou dialect has seven tones, and Cantonese has a whopping nine.

But do we have any information on where these tones come from? After all, the Qieyun traced back to the early 7th century, right around when Middle Chinese was spoken. Although not meant to be a guide on how to speak Middle Chinese, but more rather, the literary classics of the era, it still offers a slight glimpse in what Chinese of that time could, and a strong emphasis on the modal meaning right there, have sounded like.

But the Qieyun has a weird quirk too. It has an indication of tone. This suggests the possibility of Middle Chinese already having a tonal system, and a potentially four-tonal system at that. But that could not have been the case forever, right? After all, Cantonese retained many consonant stops not seen in Mandarin today, in addition to its relatively colossal tone inventory.

This theoretically could be the case. Chinese could have had more final consonants that lead towards the formation of tones we hear today. Early Middle Chinese had three phonemic tones in most syllables, but no distinctions for those ending in the consonants “k”, “t”, and “p”. When these consonants were lost, they became the checked fourth tone. While this theory sounds simple, it could be way too simple. Chinese today still uses the “n” and “ng” final consonants, and syllables could adopt any of the four tones. Of course sound changes would have happened along the way, but this proposition would have suggested that there would be a time when Middle Chinese would have had a tone where neither “n” nor “ng” would be the final consonant.

A more plausible theory would provide explanations for all four phonemic tones. After all, it is widely held today that Old Chinese might not have phonemic tones at all, instead, it had various consonant stops, including the glottal stop, “k”, “t’, “p”, and “s”. As these were lost, they would form tones. This forms the principle behind tonogenesis here- that tone arises as a result of loss or merger of consonants.

It was theorised that the rising and falling tones in Chinese came from the merger of the /s/ and glottal stop sound into /h/, and later, its eventual loss. This glottal stop could still be found in related languages like Hakka and Min.

The entering tone still originated as mentioned earlier, while the level tone arose from syllables where the consonants at the end of syllables were not /s/, glottal stop, /k/, /t/, or /p/. There are also other notable sound changes over time, as tone splits occurred based on the voicing of the initial consonant.

These after all, still are theories. Exactly when or why tonogenesis happens so widely in China remains a rather interesting linguistic mystery. We still see Sino-Tibetan languages that retain the consonants that were lost in Chinese, yet have a comparable number of tones, like Cantonese, which has “k”, “t”, “p”, and “m” final consonants just to name some, in addition to its nine tones. The mechanisms through which tone arises in languages like these continue to interest me, and hopefully, I have helped shed some light on how or roughly when Chinese became tonal.