Some days I think I don't know what I'm doing when I mess around with formant
analysis. Other days I'm sure I don't.
For what it's worth, though, I've done some analysis - using Praat - on Mark
Barratt's vowels in his
rendition of "Can a cat man a catamaran?". The results comprise a plot of the
formants of each "a" vowel in "can", first "cat", "man", second "cat", and "ran"; and a
vowel chart using the presentation format recommended by Ladefoged in his A Course in
The outstanding feature of the vowels is that they all have a substantial amount
of glide. This shows how inaccurate it is to refer to a vowel as representing
a small area of the vowel quadrilateral. That is, we can say that someone's
[&] is a low, front, unrounded vowel, but in actual speech, the vowel that's heard may
glide from very low to near mid, as does Mark's "a" in "man", or from far front to
central, as does the "a" in his second "cat".
A large amount of glide is typically found in vowels in running speech. This
assertion is backed up by the vowel examples in William Labov's Principles of Linguistic
Change Volume 1: Internal Factors. Non-gliding vowels in English
speech appear to be quite rare.