I am not sure how modals work. I just learning the language and at first thought about posting this to lojban-beginners, but was adviced on #lojban to post it here, as the argument could be non-trivial.
The firtst two alternatives I see are:
M1. Modals add additional places to predicate, creating new predicate related to basic, but with different place-structure:
5.1) mi viska do fi'o kanla [fe'u] le zunle
Basic predicate "x
1 sees/views/perceives visually x
2 under conditions x
3." modified into "x
1 sees/views/perceives visually x
2 under conditions x
3, the eye being x
m"
6.1) mi viska do sepi'o le zunle kanla
The same predicate is modified into "x
1 sees/views/perceives visually x
2 under conditions x
3, using x
m as a tool"
M2. Modals introduce additional predicates, linked to the main one... somehow. Example 6.1) would be interpreted as:
"P
1: x
1 sees/views/perceives visually x
2 under conditions x
3.
P
2:using x
1 uses/employs x
m [tool, apparatus, machine, agent, acting entity, material] for purpose (du'u) P
1."
It is difficult to interpret example 5.1) in such a way, as {kanla} does not offer place to link subordinate predicate directly, as is the case with the third place of {pilno}: "x
1 is a/the eye [body-part] of x
2" — here, we only can connect x
2 of {kanla} with x
1 of {viska}, but not with whole P
1
Modal connectives seem to support this alternative:
7.1) le spati cu banro ri'a le nu do djacu dunda fi le spati
7.5) do djacu dunda fi le spati seri'a le nu ri banro
7.6) le nu do djacu dunda fi le spati cu rinka le nu le spati cu banro
7.7) le spati cu banro .iri'abo do djacu dunda fi le spati
All sentences under this interpretation have same predicate structure:
P
1 :x
1 grows/expands [an increasing development] to size/into form {zo'e}
from {zo'e}
.
P
2: {nu} P
1 (event/state) effects/physically causes effect {nu} P
3 (event/state) under conditions {zo'e}.
P
3: y
1 [donor] (water-type of) gives/donates gift/present {zo'e}
to recipient/beneficiary y
3 [without payment/exchange].
but differ on which predicates are claimed and which are held as abstractions.
(By the way, am I right in understanding what only those modals which have short BAI form could be used in connectives?)
There could be third alternative, or at least, additional factor to consider that a recent discussion on #lojban touched. If multiple modals are present in the same statement, then the order of their appearance could matter, as they modify the main predicate one-by-one, creating scopes:
L1) {se pi'o lo forca ka'ai lo gerku mi citka}
Using a fork as a tool (accompanied by a dog (I am eating))
L2) {ka'ai lo gerku se pi'o lo forca mi citka}
Accompanied by a dog (using a fork as a tool (I am eating))
This view seem to be more easily combined with the first of my alternatives above, as ka'ai transforms {citka} "x
1 eats x
2" into "x
1 eats x
2, accompanied by x
3.", but still difference between full sencences is unclear, as they have same places. Second alternative, though, preserve the structure of scopes fully — but forces to introduce compound predicates:
"P
1 : x
1 eats/ingests/consumes (transitive verb) x
2.
P
2: x
1 is with/accompanies/is a companion of x
m1, in state/condition/enterprise P
1 (event/state).
P
1&2: P
1 & P
2P
3:using x
1 uses/employs x
m2 [tool, apparatus, machine, agent, acting entity, material] for purpose (du'u) P
1&2."
(and vice versa for L2) )
I have to admit what view M2 is influenced by how modals are handled in gua spì and Tòaq Dzũ, but also seem to be supported, as I mentioned before, by how modal connectives work in lojban. So what do you think, which view is more correct? Or is there some other interpretation?