Cursor Nictans dixit:
We have to forgo the usual Latin rule about counting verse-final syllables as heavy (which isn’t the practice in Japanese anyway).
That's not true. The usual rule is more like it doesn't matter which quantity the last syllable has. This may lead you to believe that it has a lengthening effect in metres like the pentametre or in iambs, where you would naturally expect the last syllable to be long, like in this one:
vade, liber, verbisque meis loca grata saluta:
contingam certe quo licet illa pede.
However, I can also think of examples where you would naturally expect the last syllable to be short, but you find it replaced by a long syllable. The easiest example would be a hexametre having a spondee in the 6th foot, but you may of course argue that spondees are not unusual replacements for dactyles, and by that token also trochees, in hexametres.
I can also think of examples in prose clausulae.
As far as I know, Cicero has a preference for finishing his sentences is dicreticusses (-v-/-v-) and dispondees (--/--), which naturally end in long syllables, but also in acatalectic (i.e. 'incomplete') dicreticusses (-v-/-v) and ditrochees (-v/-v), which would naturally end in short syllables. However, for all of those clausulae, you may replace the last syllable with a syllable of the other quantity (so they're essentially anceps ... or ancipites for that matter)
a well known example:
Quo usque tandem abutere, Catilina, patientia / nostra?
acatalic dicreticus -v- / -
- (- instead of v)
quam diu etiam furor iste tuus nos eludet? quem ad finem sese effrenata iactabit au/dacia?
dicreticus -v- / -v
v (v instead of -)
Nihilne te nocturnum praesidium Palati, nihil urbis vigiliae, nihil timor populi, nihil concursus bonorum omnium, nihil hic munitissimus habendi senatus locus, nihil horum ora voltusque mo/verunt?
acatalectic dicreticus -v- / -
- (- instead of v)
To sum it up: A syllable at the end of a verse or sentence may either be long or short and there is no lengthening or shortening effect ... at least in open syllables.
However, with closed syllables I *feel* (i.e. I can't really prove it) that the end of a verse may have a lengthening effect because the stop it causes feels quite similar to a consonant following up. I also base that feeling on the fact that a) verse endings do not cause elisions and b) even caesurae (weaker pauses) may have such an effect. Therefore I'd say that a closed syllable that is short by nature may, if it is followed by only one consonant (e.g.
tenet) be regarded as either long or short at the end of a verse.
In some very few cases caesurae may even lengthen short, open syllables. Maybe you'd even get away with counting a short final open syllable as long if you desperately need that for your haiku