krtisfranks wrote:
> I actually think that "ro'oi" should not require and implication like
> that at all. It might require implication, but it is not necessarily
> structured in any of those ways. "∀x ∈ A, (P(x))" just means "x ∈ A ⇒
> P(x)"; if "A" is omitted, then the universal set/set of discourse is
meant
> (where "set" can be replaced by "category", "object", or some other
> structure, depending on context).
>

This is where one of xorlo's weirdnesses shows itself. Single-predicate
ro'oi is supposed to mean that in a domain of individuals, all
combinations of them satisfy the predicate. Treating two-predicate ro'oi
the same, just with the domain stated explicitly, would result in
unexpected meanings when both of the predicates are non-distributive.
Take:

> ro'oi tavlysi'u cu nelcysi'u ja xebnysi'u

Say the domain contains the individuals A, B, C, D, E, of which A and B
talk to each other, and C and D do too. Under the reading with two
predicates and implication, this sentence would mean that:

- A and B like or hate each other, and
- C and D like or hate each other.

Taking the tavlysi'u to specify a domain for a simple
> ro'oi da nelcysi'u ja xebnysi'u
on the other hand, would give a meaning of:

- A likes or hates each other (sic), and
- A and B like or hate each other, and
- A, B and C like or hate each other, and so on for every combination of
A, B, C and D.

So I think only the all-predicate version (your second option) would make
sense here.