Definition of a vowel from dictionary.com:
Air flow itself can make vowels without the adduction of the vocal folds. We call this whispering. But for sung sounds, the vocal folds must come together to vibrate, hence providing an extra step beyond mere “air flow”. Given that the folds must come together to vibrate, let’s focus on what happens above the level of the folds to make the various vowels.
The different vowels are made by changing the shape of the vocal tract. On one pitch we can sing many different vowels, each having a different shape of pharynx, mouth, and mouth opening. In other words, we change the shape of the instrument for each vowel.
On a clarinet or a flute, we have an instrument with a basic shape that becomes shorter or longer as pitches change, but it will never suddenly change its tubing diameter or go from a perfectly round cylinder to something more oval in cross section, or from a thicker tube wall to a thinner one, even when the player adjusts his embouchure or blowing pattern to vary the timbre. However, the human vocal tract exhibits large changes going from one vowel to the next, and/or when the singer attempts a different timbre on the same vowel (classical compared to jazz vocalizing, for example).
It seems to me that crossing over from one style to another is the equivalent of playing the same instrument with a different kind of tone quality. Listen to examples of classical saxophone or trumpet compared to blues and it’s very apparent. But changing vowels (as required by sung lyrics) is more like playing a slightly different instrument for each vowel. The instrument is reformed for every vowel. The space in front of and behind the arch of the tongue changes, the mouth shape and opening change, and the position of the soft palate changes. There is also evidence that parts of the pharynx just above the larynx change, although these are considered to be changes that have more to do with timbre/genre than with vowel formation.
The big question becomes: How can we make the vowels equally resonant and still have distinct vowels? Won’t an /a/ always be more brilliant than a /u/? How far to go? How to make /u/ and /o/ bright enough and /i/ and /e/ mellow enough, but still keep clear boundaries that make lyrics intelligible? There are many instances in many genres of singers choosing a certain kind of instrumental sound over lyrics, and also many instances of singers choosing lyric intelligibility over instrumental consistency. Choosing when to do which is part of vocal artistry.