History: construal of antecedents of sumti anaphors

Source of version: 2 (current)

There are two counterproposals here. The first is by me (= And), who posted it in the belief that Jorge agreed with it, and the second is a counterproposal by Jorge. I (= And) think Jorge's proposal is better, though; it requires less forethought. (BTW, it hasn't yet been established how the two proposals would apply to sumti anaphors based on ''nei/no'a''.) ((--And)) ---- If the antecedent of a sumti anaphor is a sumti that is quantified in situ, then the anaphor repeats the quantification: ''le prenu ku ri prami'' "Each of the people loves each of the (same) people" ''lo prenu ku ri prami'' "Somebody loves somebody" To have the antecedent be just a bound variable, the antecedent sumti mustn't be quantified in situ: ''le prenu goi ko'a zo'u ko'a ri prami'' "Each of the people loves themself" ''da poi ke'a prenu zo'u da ri prami'' ~ ''lo prenu ku goi ko'a zo'u ko'a ri prami'' "Somebody loves themself" [[mi'e And] ---- I prefer a different convention. If the sumti anaphor is not quantified, then it is just like a bound variable: ''le prenu ku ri prami'' "Each of the people loves themself" ''lo prenu ku ri prami'' "Somebody loves themself" If the anaphor is quantified, then it is a new binding, restricted to the same set as the antecedent: ''le prenu ku ro ri prami'' "Each of the people loves each of the (same) people" ''lo prenu ku su'o ri prami'' "Somebody loves somebody" ''le prenu ku su'o ri prami'' "Each of the people loves somebody" ''Shouldn't this be "Each of the people loves at least one of them (the same people)"?'' Yes. --((xorxes)) ''lo prenu ku ro ri prami'' "Somebody loves everybody" mi'e ((xorxes))