Ask Mr Linguistics Person

Dear Mr Linguistics Person, A friend and I recently were talking about the remarkable cross-linguistic diversity in animal onomatopoeia. During the course of the discussion, it occurred to me that if the noises associated with animals are attempts to mimic them (admittedly in some sort of iconic manner), then the more native phonemes one’s language has, the more effectively one should be able to approximate any given animal. But this is clearly not the whole story, as vowel and consonant inventories are presumably of very different utility; would a language like Danish, with 12 distinctions in vowel quality, plus length, plus allophones with /rV/ and /Vr/ be able to represent certain sorts of animal noises better than, say, a language like Ubykh, with its myriad of place and manner distinctions in its consonant series and utter lack of vowels? And are both better at approximating animal noises than a phonemically impoverished language like Rotokas? It seems like no one is more likely to know than Mr Linguistics Person. Thanks, Ryan P.

Ryan, your intuitions in this matter are leading you in the right direction, but the truth is even more complex. In addition to considering consonant and vowel inventories and phonotactic complexity, you also need to consider the sociolinguistic factor of verbal conservatism—which is a measure of how open the members of a given culturolinguistic milieu are to non-standard–but–useful phonemic representations. Variation in this dimension at the idiomilieu level is more obvious—we all know people who would not make a verbal “sound effect” to save their lives, and those who cannot tell a story about a dog barking or yelping, or the unhealthy noises their car makes, without mimicking those sounds as accurately as possible. Similarly, at the societal level, some sounds are acceptable as “onomatopoeia”, while others cross the line into “imitation”.

The relationships aren’t too difficult to tease out with the right tools. Below I’ve plotted basic, widely-available information on consonant inventories, vowel inventories, syllable complexity, verbal conservatism, and onomatopoeticality for one hundred randomly-selected languages. The graph is a standard inverted logarithmic cumulative plot with dependent variable substrate. With this elucidative data visualization, the relationship among the variables is quite clear.

Hope that helps.
Mr Linguistics Person

	Linguistics Nerd Camp—Bethany Carlson
	Ode to Jungftak, et al.—Nihila R. Tikel
	SpecGram Vol CLXI, No 4 Contents