History: Relative Clause Veridical Quantification Gotcha

Source of version: 2 (current)

Relative clauses can attach to the ((selbri)) of a ((sumti)), or to the entire ((sumti)): *mi viska le {prenu poi klama} ku *mi viska le {prenu ku} poi klama The default is the first; and it doesn't normally any make any difference. But if you're using a ((veridical)) determiner (e.g. ''lo''), and a ((non-veridical)) relativiser (''noi''), note that the claim is by default made of the ((selbri)) and not of the ((sumti)). So *mi zvati re lo {muno merjecta noi snanu} means ''I've been to 2 out of the {50 U.S. states, which are in the south}'', and __not__ ''I've been to {2 out of the 50 U.S. states}, which are in the south.'' What you should say is *mi zvati {re lo muno merjecta ku} noi snanu Things get even worse without overt quantifiers. ((The Complete Lojban Language)), [http://www.lojban.org/files/reference-grammar/chap8.html|Chapter 8] gives example (6.8): *lo prenu noi blabi [[ku] cu klama le zarci This means not ''A person, who is white, went to the store'', but ''One out of {all people, who incidentally are white} went to the store'' -- namely, it assumes that all people are white! Sagely, therefore, does ((The Book)) state: ;:The safe strategy, therefore, is to always use ''ku'' when attaching a ''noi'' relative clause to a ''lo'' descriptor. Otherwise we may end up claiming far too much.