(1) Rule #2 describes particles that do not have enough phonological weight to carry an accent of their own, so that they combine, for pronunciation purposes, with the word that follows them.
Rule #10 then describes particles that display the identical behavior, except that they combine with the preceding word instead of the following word.
To me, those would both be called "clitics", the term I know for a word which is phonologically dependent on another word while being grammatically independent. You name the second group as "enclitics" and have no name for the first group. Do you see them as being different phenomena? Why treat them so separately?
(Actually, #2 appears to include #10 in that it mentions "a word that has two accents at different places", while the only example of such a thing arises from rule #10.)
(2) I wanted to think of the accent location as being sensitive to morae rather than syllables. But the examples clearly show that while a long final syllable will drag an antepenultimate accent forward, a long penultimate syllable will not. Might you be able to talk about why that is?
Thanks for these questions.
(1) The cases under rule #2 are a mixture of proclitics and common words (such as the two conjunctions and the adverb ou). Most proclitics (including most prepositions, and most forms of the definite article) in fact do keep their accent, so a general rule can't be given. Since enclitics do follow set rules, which are much more complex than simply not having an accent, they are given their proper treatment separately in the last (and most tricky) of the rules!
(2) Analysis by morae, whereby a short syllable counts as one mora, and a long syllable as two, generally allows the rule that an acute cannot go back more than four morae, and a circumflex more than three morae - but you still have to deal with the fact that, as you say, a final long restricts an acute to three morae from the end, and face the exception given in n.13. So we decided to avoid it entirely!
(1) Rule #2 describes particles that do not have enough phonological weight to carry an accent of their own, so that they combine, for pronunciation purposes, with the word that follows them.
Rule #10 then describes particles that display the identical behavior, except that they combine with the preceding word instead of the following word.
To me, those would both be called "clitics", the term I know for a word which is phonologically dependent on another word while being grammatically independent. You name the second group as "enclitics" and have no name for the first group. Do you see them as being different phenomena? Why treat them so separately?
(Actually, #2 appears to include #10 in that it mentions "a word that has two accents at different places", while the only example of such a thing arises from rule #10.)
(2) I wanted to think of the accent location as being sensitive to morae rather than syllables. But the examples clearly show that while a long final syllable will drag an antepenultimate accent forward, a long penultimate syllable will not. Might you be able to talk about why that is?