"If we're looking for sounds that would be PERCEIVED as something other than vowels or consonsonants or something along that continuum -- that's an issue about the perceivers, isn't it? If they've been trained that to be meaningful, a sound must be classified as v, c, or in between -- then won't anything that might be meaningful be stuck into one of those categories, whether it physiologically fits the physiological definition or not?"
Yes. Terran linguists listening to the speech of native speakers of an ET language are going to expect to hear vowels and consonants because that's what they've been trained to hear, and are going to sort the sounds they hear into those two categories for that reason. Only after the U.S. Corps of Linguists (USCOL) had accumulated a large database of ET sound-based languages that included vowels, consonants, and "something else" would it be possible to train them to identify and analyze that "something else." And my conviction is that that would take a very long time to happen.
And here's the next paragraph of the comment:
"Are we looking for sounds from the vocal tract that would be so different physiologically/phonetically that they COULD NOT be fitted into those v-c categories, even by a sort of legal fiction? But would somehow be clearly meaningful so that they COULD NOT simply be disregarded or somehow marginalized?"
Yes again. My Brethandi ETs -- because their anatomy is very different from the anatomy of the Terran cattle they so closely resemble to the casual eye -- are able to speak in a fashion comparable to Terran speech, although they of course have distinctive accents. [I knew that. So I did a cognitive SHAZAM-leap and took it for granted that you would know it too. Sheesh.] And my question was serious. Supposing one or more of the Brethandi languages was composed of three meaningful classes of sounds -- vowels, consonants, and something else -- then what, I wanted to know, could that something else possibly be?
One possibility turned up in a comment from
"Consonants, vowels, and rests. As in music. The rhythmic and intentional interruption of consonants and vowels to modify their meaning."
That option -- musical rests -- is the one I used in my USCOL story "Honor Is Golden," published in Analog. [Not online anywhere, so far as I know.] It worked, to my satisfaction and my editor's, although I got slammed for it in a review. My linguists weren't able to isolate the rest-phonemes or work with them, but they were able to establish communication. Which was their primary goal.
I hope this clarifies things just a tad. If it doesn't, let me know and I'll try again.

