[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lojban] Re: {tu'e...tu'u} in NU



Le mercredi 15 juillet 2015 04:26:50 UTC, Spheniscine (la zipcpi) a écrit :
One problem that crops up now and then concerns how to include more than one bridi within a NU clause. An example is my translation of Deuteronomy 8:12~14, which I have rendered:

.i tezu'ebo rivbi va'o lo nu do mo'u citka jebo cu tolxagji je cu zbasu lo zabna zdani je cu xabju ri
ju'ei loi cagda'u be do ferti se panzi ju'ei loi rijno joi loi solji vu'o po do zei'a sormei ju'ei loi ro se ponse be do zei'asai sormei kei
lo nu lo menli be do co'a se'ijgi ju'ei do tolmo'i tu'a la .iauex. ku noi cevni do je noi vimcu do la .misr. ku no'u le tutra pe lo ka bapse'u

(note, uses simplified connectives, the experimental tag {zei'a} which is basically {fi'o te zenba}: {zei'a sormei} means "increasingly many" / "to increase in number", and {ju'ei}, which shall be discussed. Yes, I'm pretty terrible with abusing experimental constructs.)

There are several solutions:
official-Lojban solution #1: use {ju'e ... gi ...}. Requires forethought, and only takes two bridi. Chaining them to allow more is impractical.

official-Lojban solution #2: close each NU clause then join it to a new abstractor sumti. e.g. {lo nu broda kei jo'u lo nu brode}. Pretty wordy.



If you need a structure similar to your translation above, official-Lojban solution #2 must be rather {lo nu broda kei ju'e nu brode}.

 
experimental-grammar solution #1: {ju'ei}; essentially acts like a tight-scope {.i} that doesn't close sub-clauses. {lo nu broda ju'ei brode}. Has the advantage of not requiring forethought, but has the disadvantage of not automatically closing any bridi tails; if one of your abstraction-bridi happens to contain a lot of nested NU or POI, they must be closed manually.

experimental-grammar solution #2: Allow {tu'e...tu'u} to act as a single bridi in NU clauses: {lo nu tu'e broda .i brode tu'u}. Does require forethought, but gives a new "level zero" for {.i} to automatically reset to until closed by {tu'u}.



La zantufa-0.2 ( http://guskant.github.io/gerna_cipra/zantufa-0.2.html ) allows a full statement in noi-clause or nu-clause. It means the following texts are parsed by la zantufa-0.2 as follows:

1. sentences connected with {i JOI} or {i (tag) bo}:

{nu broda i ju'e brode}
(CU [nu {<CU (¹broda VAU¹)> <i ju'e (¹CU [brode VAU]¹)>} KEI] VAU) 
{nu broda i bo brode}
(CU [nu {<CU (¹broda VAU¹)> <i bo (¹CU [brode VAU]¹)>} KEI] VAU) 
{nu broda i ba bo brode}
(CU [nu {<CU (¹broda VAU¹)> <i ba bo (¹CU [brode VAU]¹)>} KEI] VAU) 

Use KEI or KUhO to cut the statement:

{nu broda kei i ju'e brode}
([CU {nu <CU (¹broda VAU¹)> kei} VAU] [i ju'e {CU <brode VAU>}]) 
{brodi ra noi broda ku'o i ju'e brode}
([CU {brodi <ra (¹noi [CU {broda VAU}] ku'o¹)> VAU}] [i ju'e {CU <brode VAU>}]) 

({brodi da} is required here because this {ju'e} connects two sentences, not a fragment and a sentence. It might be allowed in the future version of zantufa, though not yet decided.)


2. sentences between {tu'e ... tu'u}:

{nu tu'e broda i brode tu'u}
(CU [nu {tu'e <(¹CU [broda VAU]¹) (¹i [CU {brode VAU}]¹)> tu'u} KEI] VAU) 


3. sentences connected with forethought connective:

{nu ju'e gi broda gi brode gi brodi gi brodo gi brodu}
(CU [nu {CU <(¹ju'e gi¹) broda (¹[gi brode] [gi brodi] [gi brodo] [gi brodu]¹) GIhI> VAU} KEI] VAU) 

(Forethought connectives of la zantufa-0.2 can connect more than three "statements (not only sentences)", and {gi'i} is used as the elidible terminator GIhI, not as GIhA. See 
http://mw.lojban.org/papri/zantufa_jonma%27o#lo_li.27erla.27i_jonma.27o
for more info.)

As 1 is allowed, your solution #1 is unnecessary, and ma'oi JUhEI in that usage will never implemented to la zantufa. 

 
And bonus official-Lojban solution #3, with an experimental shorthand that doesn't require new grammar: {tu'a la'e lu broda .i brode li'u}, with {tu'ai} (selma'o LU) being shorthand for {tu'a la'e lu}. Useful, but may be semantically vague, given lack of LO NU qualification, and {tu'a} possibly implying missing information.

Bonus problem: A related problem is how to make several sentences share a sumtcita-term. e.g: *{ca lo nu mi stuvi'e le sralygu'e kei tu'e lo zabna cu fasnu .i lo mabla cu fasnu tu'u}. {ju'e...gi} also works here, but retains the chaining problem, while I don't know if any of the other listed solutions work.



Use {zo'u}:

{ca lo nu mi stuvi'e le sralygu'e zo'u tu'e lo zabna cu fasnu .i lo mabla cu fasnu tu'u}

It is parsed even by the official parser:

([{ca <lo (¹nu [mi {CU <stuvi'e (²le sralygu'e KU²) VAU>}] KEI¹) KU>} zo'u] [tu'e {<(¹lo zabna KU¹) (¹cu [fasnu VAU]¹)> <i (¹lo mabla KU¹) (¹cu [fasnu VAU]¹)>} tu'u]) 

zo'u-clause spans over a statement, that is to say, sentences connected with {i JA/JOI}, {i (tag) bo}, {JOI/tag gi ... gi ...} or {GA ... gi ...}. These connectives can be used instead of {tu'e ... tu'u} in many cases. For example,

{ca lo nu mi stuvi'e le sralygu'e zo'u lo zabna cu fasnu .ibo lo mabla cu fasnu}

is parsed by the official parser as follows:

([{ca <lo (¹nu [{mi CU} {stuvi'e <le sralygu'e KU> VAU}] KEI¹) KU>} zo'u] [{<lo zabna KU> cu} {fasnu VAU}] [i bo {lo mabla KU} cu] [fasnu VAU]) 

 

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.