The Origin of Tonal Consonants in Native American Languages

Iain Paul Anderson
Junior Data Scientist (FTC)
Munich University Deep Diachronic Linguistics Experiment

While preparing data from a sample of Native American languages for mass lexical comparison, I noticed a curious feature of the phonology of these languages. We normally expect tone to occur on vowels, but a large number of the languages in the sample contained consonants marked for tone. It was always the same four consonants on which tonal marking occurred—the palatal stop and approximant, and the alveolar fricatives, and they always contrasted rising tone against unmarked—no other tone was marked on these consonants, nor were any other consonants marked for tone.

Since this curious feature is found throughout the Americas, it must have been present in the language of the first inhabitants of the continent, and understanding its origin will tell us something important about Proto-American. We know that the first settlers came via the Beringia land bridge, and were therefore likely to have spoken a tonal language already, as these are pervasive in Eastern Asia.

i → j is a common enough sound change, but we would normally expect it to neutralize tonal distinctions. However, there must have been some context in which the contrast between ǐ and other tones was so salient that the distinction was preserved, thus giving rise to ǰ. Professors Thanneven and Lemma-Lunn assure me that by crunching enough data, we might even be able to work out what it was. The other tonal consonants would then have arisen by palatalization, as kǰ → č, sǰ → ʃ̌, zǰ → ǯ. Palatoalveolars then merged with alveolars, giving ʃ̌ → š, ǯ → ž. We can see that palatoalveolars are entirely absent in all languages with tonal consonants.

There is a conspicuous gap in the consonants produced by this mechanism. We might expect tǰ → t͜ʃ̌ → tˇ, dǰ → d͜ǯ → dˇ, nǰ → ɲ̌ → ň, but these are nowhere seen. Whether these consonants resisted palatalisation or whether they later lost the tonal contrast is a matter for further research.

Another peculiar consonant that occurs in many of these languages is ñ. Unfortunately, since my fixed term contract is coming to an end (and didn’t turn out to be self-employment on a lucrative daily rate), investigation of how it’s possible to nasalize a consonant that’s already nasal to start with will have to wait for another time.

	Linguimericks—Book ५३
	A Complainogenic Model of Snoring—Otto Rhine O’Leary-Ngoloji and S. Lee Papnea
	SpecGram Vol CLXXXII, No 2 Contents