I agree with you that we will never know for certain just how William Vickers and his contemporaries played these tunes, though I am sure scholars will exercise themselves with this. I can't help feeling however that the beaming of quavers in the Vickers MS is an attempt to indicate where the stress lies in the bar. Of course some of Vickers' contemporaries may have played the tunes differently. The Dusty Miller is interesting because while the first two strains are beamed in threes, the third strain is beamed in twos/fours.
Anahata had an interesting suggestion of an accompaniment pattern of
"A common 2 bar rhythm sequence , for instance: 2 2 2 | 1 2 1 2 |"
The crotchet-minim pattern implied in the second bar is quite common in renaissance music. Galliards in particular show a variety of rhythmic shifts. The underlying pattern is usually minim-crotchet with shifts to crotchet-minim or to three minims. I wonder if, and this is pure speculation, what we have with some 3/2 hornpipes is a relic of a rhythmic structure which was once common but has elsewhere disappeared. OTOH, other 3/2 hornpipes are clearly three in a bar throughout.
Of course at the end of the day, it is a matter of personal judgment when playing how to place the stress for best musical effect.
To respond to Dazbo, the underlying three in a bar is quite slow although the tunes often sound much quicker because the basic minim pulse is usually broken up into shorter length notes. There is a Purcell Hornpipe I play with a recorder group where the top line is largely quavers and semi-quavers while the lower parts are an accompaniment in mostly minims and crotchets. Playing one of the lower parts is surprisingly tricky because you are tempted not to hold the notes on long enough, as the underlying slow three is masked by all the activity on the top line. I usually cheat and count the bars in six in my head. That way I keep time better.