[lojban-beginners] Re: Tenses in abstractions

[Ian Johnson <blindbravado@gmail.com>]

2010/2/25 Jorge Llambías <jjllambias@gmail.com>:
> On Thu, Feb 25, 2010 at 12:58 PM, Ian Johnson <blindbravado@gmail.com> wrote:
>> I tried translating "When did you go to sleep last night?" I know
>> about {sumti tcita} for this (ti'u), but I'm not sure if you can use
>> them to fill 3 of tcika's places instead of just 2, so I did:
>> ma lenu do pu co'a sipna cu tcika le prulamcte (be le cabdei)
>>
>> I have four questions:
>> 1. Just in general, is that structured other-than-terribly?
>
> The "pu" is a bit redundant, since "last night" already shows that it
> was in the past.
>
> I would say:
>
>   ti'u ma do co'a sipna ca lo prulamcte
>   "At what time did you fall asleep last night?"

I like that form; as I mentioned I would've used ti'u if I could think
of a more elegant way to do it than {ti'u} and {teti'u} (at that
point, as I was saying, you might as well just use tcika, in my
opinion). It feels less clunky and is of course shorter. I suspect
that I've started leaning on {nu} structures more than I should,
seeing this and some of the past responses to my questions.


>> 4. As in the title, how does time work in abstractions which are
>> assigned tenses outside the abstraction? Here the event is assigned a
>> time in the past by the x3 place of {tcika}; so what happens to the
>> {pu}? Is it now before the assigned time of the abstraction (meaning
>> that in this case the {pu} is not what is wanted)? Is it relative to
>> the main {bridi}?
>
> {lo nu do pu co'a sipna} descibes a start of sleeping that happened
> before something. It doesn't say before what exactly, but the usual
> reference is the time of the utterance. Since you are already saying
> by other means that this starting event took place last night, the
> "pu" doesn't really add much.

OK, I see. I think what I was thinking, while a somewhat interesting
concept (an abstraction has tense(s) associated with it, and tense(s)
within that abstraction are thus relative to that time, rather than to
the speaker), I suspect that it might have issues with recursion (a
problem that is mentioned in LfB a few times, such as with using {ri}
in relative clauses) and would thus be a problematic feature in
general.

mu'omi'e latros.