But saying that the name of the character “&” was pronounced “and per se and” is like saying that the name of the character “Z” is pronounced “and zee” or “and zed”.
One could say whatever one wants but I think that in this case it’s actually a conjunction and I doubt anybody has ever said “I per se I” (Edited to add: actually that makes sense - for the letter i - but in a different context: spelling and not alphabet recitation. “c, a, r : car”, “i : I”. _letter_ (per se) : _word_)
Re "I doubt anybody has ever said I per se I" (in the context of alphabet recitation) may I draw your attention to this article[0]:
> Up until as recently as the mid 1900s, it was standard practice when reciting the alphabet to use the Latin phrase per se (literally “by itself”) to differentiate between individual letters of the alphabet—like A, I, and O—and single-character homographic words—like a, I and O.
> So the letter A would be read as “A per se A”, to ensure it was distinguished from the indefinite article a. The letter I, similarly, would be “I per se I” to differentiate it from the pronoun I. And the letter O would be “O per se O” to differentiate it from the interjection O!
How is it like saying "and zee"? The article pretty clearly outlines the logic here, and this would not be like that. If anything, it'd be "zee per se zee".
Because the alphabet was “a, b, …, x, y, z and &” (pronounced “… why zed and per se and”) and now it’s “a, b, …, x, y and z” (pronounced “… why and zed”).
You're missing that "& per se &" would still have been called that even if it wasn't the last item in the list. If the order was "...y, & and z" you'd pronounce it "why and per se and and zed".
But saying that the name of the character “&” was pronounced “and per se and” is like saying that the name of the character “Z” is pronounced “and zee” or “and zed”.