Something else. The old rules that they teach in school about long and short vowels is a gross generalization (do they even still teach that?). These words have a “short” a but they are all pronounced differently:
The whole long/short thing is a leftover of an idiotic attempt to forcefit English into a Latin mold. Generally speaking English does not use the long/short distinction for any phonemic distinction. There are two exceptions I am aware of and possibly a few more: rider (long) vs. writer (short) and ladder (long) vs. latter (short). I have read that English uses two dozen distinct vowel sounds and I don’t they all have names. And the rules, if any, are extremely dialect dependent. For example, I just had occasion to read something about Stan and Jan Berenstain. I imagine that for most of you, Stan and Jan rhyme. But not for me. And not for them either since they, like me, grew up in West Philadelphia. But how did I even know that the two a’s were different? I guess because the a in Stanley is different from the one in Janet. But my internal grammar knows things I don’t know.
According to Wikipedia, dictionary.com, Miriam-Webster, and MacMillan, the final vowel is “ee”, although some of them give “eye” as an alternate pronunciation. It’s a general problem in pronouncing Latin words in English, since there are several variant pronunciation systems (Classical, Ecclesiastical, Traditional English, etc). In ancient Latin it would have been pronounced “ee”.
Plus I know you’re in the U.K., but even looking up various British speaker readings of Yeats’s famous poem “Lapus LazulI,” they’ve all been pronounced with an “ee” sound. Plus within the poem, while it is not in strict rhyme, the “ee” pronunciation is strongly suggested, being paired with the word “longetivity” a couple lines down.
On this side of the water, there’s no distinction in the vowel length that I can tell in those examples. And to my ears, some Americans tend to pronounce some “t” sounds much closer to “d” than we would anyway, which would make them sound even closer to each other.
I absolutely agree about standardised rules, but generally speaking, [vowel]+[single consonant]+[e] indicates, if not a longer vowel sound, then at least a rounder one, and very different from [vowel]+[double consonant]+[e], where the vowel sounds flatter and crisper - and in other Germanic languages as well.
It’s pretty much all Americans that pronounce “t” similar to “d” in intervocalic situations like “petal” vs “pedal” and “ladder” vs “latter.” It’s not the usual “d” sound, but rather an intervocalic flap. We’ll do it in words like “eating,” “attic,” and “Saturday,” but not works like “mad” or “fifteen” or “anyday.” There’s a list of “rules” here. I put “rules” in quotes, because it’s not something that we formally learn. It’s just a feature of our accent that we pick up naturally. It sounds really weird to me when somebody enunciates a “t” in these situations but otherwise has an American accent. (I actually came across one person last week who did this with a specific word, and it was driving me nuts. I can’t remember the word, unfortunately, and it was only with that particular word she would pronounce the intervocalic “t” as a “t” rather than a flap.)
In my Great Lakes dialect, they are two distinct sounds. I imagine your dialect is probably part of the cot-caught merger. Do those two words sound the same to you? The majority of American dialects (about 60%) distinguish those two sounds. To us not in the merger, it’s the difference between an “ah” sound and an “aw” sound. But if you’re part of the merger, I imagine that description doesn’t mean anything to you.