The TTC evidently has a computer-generated voice for some system announcements. (Or perhaps the man on the mic today has a bummer of an accent).
But this is the first time I've ever heard a voice over a PA pronounce Spadina the way I understand the man being referenced pronounced his own name. (Spa-DEE-na rather than Spa-DYE-na)
Today, incidentally
(
Read more... )
I was taught in school that a vowel by itself was short, but a vowel with a second vowel in the same syllable was long:
"kit" versus "kite"
"kitten" has two "t"s to delineate the syllables, so the first "i" is forced to be short. If it was spelled "kiten" the "i" would be long.
Looking at the name "Spadina" on would expect the syllables to be divided either Spa-Dina or Spad-Ina. In either case, the "i" is coupled with the "a" (a second vowel) and so the assumption that "i" should be long is not unwarrented.
However, as I understand it, Ojibway (like Italian) uses only one sound per vowel (although they have a whole wack of both nasal vowels and dipthongs), and "i" in Ojibway is short, like "kit" or "pit". So really, the pronounciation should be Spa-DIN-uh, making both SpaDYNa and SpaDEENa wrong. But Spa-DIN-uh would would sound flat to most English ears.
So the question really becomes how far has the noun been assimilated into English, and how should standard syllabification apply?
[Side note: Another name which I've gotten in arguments about is "Islington". In Toronto we divide the syllables: Is-Ling-Ton. When I lived in a town in southern Missouri, they also had an Islington St., but they divided it Isl-Ing-Ton and pronouced the "Isl" the same "isle," and insisted they were right to do so, 'becuase you say eye-land, not is-land'.
Reply
language - how it evolves & corrupts is endlessly fascinating
thank you for more insights :)
Reply
Leave a comment