Ancient Greek, free word order and the same FA two times in a bridi

[Gleki Arxokuna]

Ancient Greek and Russian have free word order. It means that instead of "Yesterday I saw a big dog" you can literally say "Big yesterday saw dog I" due to case markers.

Lojban currently can't split sumti apart.

{mi viska lo barda gerku} = {fe lo barda ku viska fe lo gerku fa mi}?

This issue remains unsettled.

ju'i jbobadna how are we going to translate Homer's magnificent style? Again distorting his work of art like it is being done when translating it to English? (Russian has other drawbacks in grammar).

Therefore I suggest that if we have the same FA more than once in a bridi it means that we can join them back with {je}. So

{fe lo barda viska fe lo gerku fa mi} = {mi viska lo barda je gerku}.

If you need to use joi, fa'u, jo'u or other connectives such partitioning is not possible.

pei?

[vitci'i]

On 08/04/2012 02:05 AM, Gleki Arxokuna wrote:
> Ancient Greek and Russian have free word order. It means that instead of
> "Yesterday I saw a big dog" you can literally say "Big yesterday saw dog I"
> due to case markers.
>
> Lojban currently can't split sumti apart.
>
> {mi viska lo barda gerku} = {fe lo barda ku viska fe lo gerku fa mi}?
> This issue remains unsettled.
> ju'i jbobadna how are we going to translate Homer's magnificent style?
> Again distorting his work of art like it is being done when translating it
> to English? (Russian has other drawbacks in grammar).
>

> *Therefore I suggest that if we have the same FA more than once in a bridi
> it means that we can join them back with {je}. So*
> *{fe lo barda viska fe lo gerku fa mi} = {mi viska lo barda je gerku}.*
> *If you need to use joi, fa'u, jo'u or other connectives such partitioning
> is not possible.*
>
>
> pei?

No language should strive to have every feature of every other language.
It's okay for lojban to be its own thing.

If a majority of the source languages have a feature that lojban lacks,
then that may indicate a problem, especially if English is in the
minority that lojban resembles.

[Luke Bergen]

This has come up a couple times before.

I don't think it's a matter of "we need to figure out how to do this because other languages do it".  I think it's more so a matter of "the grammar allows this form (fe lo barda cu broda fe lo gerku), we should probably decide just what it means".

I like what you're suggesting Gleki, but I disagree about the connector.  In cases where a word is kind of being magicked in, I would almost always prefer that that word be as vague as possible so as to include as broad of a scope of meaning as is possible.  

So I'd rather it be {ju'e}.

--
You received this message because you are subscribed to the Google Groups "lojban" group.

To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.

[Gleki Arxokuna]

On Saturday, August 4, 2012 6:12:18 PM UTC+4, pafcribe wrote:

This has come up a couple times before.

I don't think it's a matter of "we need to figure out how to do this because other languages do it".  I think it's more so a matter of "the grammar allows this form (fe lo barda cu broda fe lo gerku), we should probably decide just what it means".

I like what you're suggesting Gleki, but I disagree about the connector.  In cases where a word is kind of being magicked in, I would almost always prefer that that word be as vague as possible so as to include as broad of a scope of meaning as is possible.  

So I'd rather it be {ju'e}.

Oh, probably. If i only knew what {ju'e} really means. It's even not mentioned in the CLL.  No examples of it's usage.

{mi viska lo barda ju'e gerku} = {mi viska lo barda .i mi viska lo gerku}?

To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.

[selpa'i]

Am 04.08.2012 09:05, schrieb Gleki Arxokuna:

Ancient Greek and Russian have free word order. It means that instead of "Yesterday I saw a big dog" you can literally say "Big yesterday saw dog I" due to case markers.

Lojban currently can't split sumti apart.

{mi viska lo barda gerku} = {fe lo barda ku viska fe lo gerku fa mi}?

Therefore I suggest that if we have the same FA more than once in a bridi it means that we can join them back with {je}. So

{fe lo barda viska fe lo gerku fa mi} = {mi viska lo barda je gerku}.

If you need to use joi, fa'u, jo'u or other connectives such partitioning is not possible.

pei?

I don't like using logical connectives for this. How about no'u?

fe lo barda fa mi viska fe lo gerku
=
mi viska lo barda no'u lo gerku
"I saw something big, which is a dog."

mu'o mi'e la selpa'i

-- 
pilno zo le xu .i lo dei bangu cu se cmene zo lojbo .e nai zo lejbo

[John E Clifford]

This quest seems based on the idea that Lojban achieves cultural neutrality by permitting a speaker of any language to speak Lojban in the style of his native language, putting in what he normally would and leaving out what he normally would not (or could not) use.  This goal has been partially realized in a number of areas, like tense/aspect, for example, but surely not in all.  I suppose it is possible in Lojban to reproduce the effect of Swahili word classes by frequent tanru all over the place, but it doesn't seem to be of much point, since the classes are semantically insignificant (Kilimanjaro is not a little hill, for example, despite the 'ki' prefix).  The same applies in other cases, mostly relevant to the parallel thread about animateness.  For free word order the case is more complex, since that affects parsing.  Lojban is said to be uniquely parsed (in a non-trivial way) and this relies ultimately on the the uniqueness of the parsing of FOL, which involves positive RHE markers (or counters).  Lojban has always had some RHE problems, most getting solved some after they appear, but new ones still turning up from time to time (and guarantee that more will not).  To split a construction whose RHE is a possible issue (a complex sumti, say) is to open a mare's nest of new problems.  We can easily reorder sumti any which way and the order of modifiers and modified within a sumti.  To go beyond that seems a needless bowing to an uncommon style at the cost of a major characteristic of the language.

---

From: selpa'i <sel...@gmx.de>
To: loj...@googlegroups.com
Sent: Saturday, August 4, 2012 9:27 AM
Subject: Re: [lojban] Ancient Greek, free word order and the same FA two times in a bridi

--

You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to loj...@googlegroups.com.

To unsubscribe from this group, send email to lojban+un...@googlegroups.com.

[Pierre Abbat]

On Saturday 04 August 2012 10:12:18 Luke Bergen wrote:
> This has come up a couple times before.
>
> I don't think it's a matter of "we need to figure out how to do this
> because other languages do it". I think it's more so a matter of "the
> grammar allows this form (fe lo barda cu broda fe lo gerku), we should
> probably decide just what it means".

Maybe it's good to figure out what this construction means, but I don't think
that, in general, it's wise to assign a meaning to every construction that
the formal grammar allows. "mi te.u do du ra'o lo gerku pe naku" I consider
to be ungrammatical on the second level, even though it is grammatical on the
first (the formal grammar).

Pierre

--
loi mintu se ckaji danlu cu jmaji

[djandus]

On Saturday, August 4, 2012 9:27:34 AM UTC-5, selpa'i wrote:

I don't like using logical connectives for this. How about no'u?

fe lo barda fa mi viska fe lo gerku
=
mi viska lo barda no'u lo gerku
"I saw something big, which is a dog."

This issue has further implications than just FA usage, IMO. (addressed below) And when I've thought about it before, I primarily remember it mainly feeling way too arbitrary a decision. Which is why I personally would prefer a vague usage, if any. (i.e. jo'u or no'i) Also personally, it feels really awkward to assume any logical connectiveness at all or direct relationship between the two instances of the FA. That is, when I see {fe lo barda fa mi viska fe lo gerku} my first impulse is that the x2 spot could be filled with either {lo barda} or {lo gerku}, and whether there is anything to be implied about that is unknown.

How this issue comes up in other grammar bits is more interesting to me, however. Particularly, how is {lo gunka be fa mi} currently parsed? If multiple FA are explicitly disallowed, then it makes sense to be able to claim that it means the same as {lo se gunka be mi}. (seems pretty useless) If multiple FA are allowed, then it would probably end up meaning something very similar to {lo gunka GOI mi} of some kind. And that opens a whole 'nother can of worms.

For me, I'm fine with never using either of these ideas, ever, ever, ever. It makes sense. I'd even be fine with having double-FA implying a si/sa/su-like erasure of the first instance. (That is, {mi lo gerku ku viska fe do} could be "I saw the dog -- no, wait, I mean I saw you.") That fits in more nicely with how GOhA pro-bridi are handled, where the repeated instances overwrite older occurances. It's extraordinarily functional for GOhA in conversation. If that were the handling for FA, I'd expect {lo gunka be fa mi} to be parsed as semantic nonsense, probably a mistake meant to be {lo gunka po'u/no'u mi}

Overall, I look at it as being a bit of a folly to assign a meaning to multiple FA, but if I were to go with anything, I'd take great care to be as consistent as possible with the other grammatical constructs. Just off of these, erasure seems most sensible to me.

mu'o mi'e djos

[iesk]

I tend to think that {broda fa ko'a fa ko'e} means more or less the same thing as {ko'a broda .i ko'e broda}, ie making two predications at once or 'putting two feet into the same boot' (or some similar expression that somebody once used). The (logical, temporal, etc.) connection between the first bridi and the second in {ko'a broda .i ko'e broda} would depend on context, wouldn't it? So, can't the meaning of 'double-FA' expressions be considered similarly context dependent?

Out of context, I would assume that both statements are to express prepositions held true (by the speaker). So, such 'double-FA' utterances may have sense or not, depending of course on the predication and the arguments involved.

{catlu fe lo melbi tsani fe lo cmalu rirxe fa ko} -> why not
{se jbena fa mi fa do la cmen.} -> probably dissonant … also, sorry for the somewhat ka'u disgusting example :/

Hm. This doesn't make much sense, does it?

·iesk·

[djandus]

On Sunday, August 5, 2012 4:38:29 AM UTC-5, iesk wrote:

I tend to think that {broda fa ko'a fa ko'e} means more or less the same thing as {ko'a broda .i ko'e broda}, 

This is also what I was imagining at first, but I didn't realize it until you said it. It occurs to me that this would probably be the most easily acceptable interpretation for dealing with the problems I brought up (i.e. how it interrelates with {be} and GOhA usage) as it leaves it up to context what relationship there is between the two statements, if any. In the case of GOhA, it would generally be an overwrite of the previous, implying some form of correction. In the case of {be} and double-FA, I think it's a somewhat silly usage, but it could make sense. 

Personally, I find the pure "erasure" sort of usage to be very useful, and any other kind of... weak. I also think logical connective implications would be explicitly wrong, as those feel completely against the GOhA overwrite usage, and it would only be a very small shortcut to far more confusing statements that are already expressible.

[Gleki Arxokuna]

doi djandus isn't it {ju'e} that pafcribe suggested?

If you agree with it we have three votes (including mine).

[Gleki Arxokuna]

On Saturday, August 4, 2012 6:27:34 PM UTC+4, selpa'i wrote:

Am 04.08.2012 09:05, schrieb Gleki Arxokuna:

Ancient Greek and Russian have free word order. It means that instead of "Yesterday I saw a big dog" you can literally say "Big yesterday saw dog I" due to case markers.

Lojban currently can't split sumti apart.

{mi viska lo barda gerku} = {fe lo barda ku viska fe lo gerku fa mi}?

Therefore I suggest that if we have the same FA more than once in a bridi it means that we can join them back with {je}. So

{fe lo barda viska fe lo gerku fa mi} = {mi viska lo barda je gerku}.

If you need to use joi, fa'u, jo'u or other connectives such partitioning is not possible.

pei?

I don't like using logical connectives for this. How about no'u?

1. Taking into account that la selpa'i cu lojbo certu

this solution has a higher priority :)

Should be put into the todo list for CLL-2.0

Other proposals are

2.ju'e/.i

3.si