- Home
- Get A Printable Dictionary
- Search Best Words
- Recent Changes
- How You Can Help
- valsi - All
- valsi - Preferred Only
- natlang - All
- natlang - Preferred Only
- Languages
- XML Export
- user Listing
- Report Bugs
- Utilities
- Status
- Help
- Admin Request
- Create Account
|
Discussion of "kancusu'oi"
[parent]
[root]
Comment #4:
Re: Why have two sumti slots?
|
la du kooi noi zo zei zei zei zei fao ku ne vou biu pei (Sun Jan 31 09:28:57 2021)
|
krtisfranks wrote: > Okay, but why not make it arbitrarily many (including possibly infinitely > many)?
Since the quantifier grammar allows at most two predicates as arguments, I hadn't thought about how they might generalize! Now that I do, I'm confused.
su'oi's expansion uses the associative ∧, and > su'oi gi [ke'a] broda gi brode gi brodi [gi'i] (forgive the grammar) would make sense as > ∃x (broda(x) ∧ brode(x) ∧ brodi(x))
but what would > ro'oi gi broda gi brode gi brodi best mean? > ?? ∀x ((broda(x) → brode(x)) → brodi(x)) > ?? ∀x (broda(x) → (brode(x) → brodi(x))) > ?? ∀x (broda(x) → (brode(x) ∧ brodi(x)))
|
-
Comment #5:
Re: Why have two sumti slots?
|
Curtis W Franks (Mon Feb 1 07:36:24 2021)
|
zozeizeizeizeifaho wrote: > krtisfranks wrote: > > Okay, but why not make it arbitrarily many (including possibly > infinitely > > many)? > > Since the quantifier grammar allows at most two predicates as arguments, I > hadn't thought about how they might generalize! Now that I do, I'm > confused. > > su'oi's expansion uses the associative ∧, and > > su'oi gi [ke'a] broda gi brode gi brodi [gi'i] > (forgive the grammar) would make sense as > > ∃x (broda(x) ∧ brode(x) ∧ brodi(x)) > > but what would > > ro'oi gi broda gi brode gi brodi > best mean? > > ?? ∀x ((broda(x) → brode(x)) → brodi(x)) > > ?? ∀x (broda(x) → (brode(x) → brodi(x))) > > ?? ∀x (broda(x) → (brode(x) ∧ brodi(x)))
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). The universal quantification is just the first portion of this clause. But, if we are to include both pieces, then I guess that I would say that "ro'oi(x, A, (P_1, P_2, P_3, ...))" would mean "x ∈ A ⇒ (P_1 (x), & P_2 (x), & P_3 (x), ..."; we can, again, omit A. A phrase "all cats" implicitly means "∀x ∈ A" where A is the set of only and all cats, too.
Alternatively, we can take P_0 to be the proposition representing membership in A, which would immediately narrow the range of relevant x values from the universal set to A. This would introduce an asymmetry in the the (now-ordered) list of propositions (because it would mean "∀x, (P_0 (x) ⇒ (P_1 (x), & P_2 (x), & P_3 (x)))"), where "∀x" means "all x in our discourse space". This would be, I think, the last option which you outlined.
I am going off how I use it and have seen it be used. We can, of course, define it in any manner which we want.
|
-
Comment #6:
Re: Why have two sumti slots?
|
la du kooi noi zo zei zei zei zei fao ku ne vou biu pei (Thu Feb 4 22:14:12 2021)
|
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.
|
-
|
|
|
|