I think I can use "kau" to express Japanese "ga" and "wa"

78 views
Skip to first unread message

.iocikun.juj.

unread,
Oct 11, 2012, 8:21:28 PM10/11/12
to lojban-b...@googlegroups.com
coi rodo

I'm sorry that I have bad English skill.
Please forgive it.

English "I go" is "watashi ga iku" or "watashi wa iku" in Japanese.

Who goes? I go. -> watashi ga iku
What do you do? I go. -> watashi wa iku

The difference is listener's knowledge.

If listener knows that someone go but doesn't know that who go, speaker say "watashi ga iku".
If listener knows that speaker do something but doesn't know that what does speaker do, speaker say "watashi wa iku".

"watashi ga iku" and "watashi wa iku" is different at listener's knowledge.

I think I can use "kau" to express the difference.

"watashi ga iku" is "mi kau klama" and
"watashi wa iku" is "mi klama kau".

I think this usage of "kau" can be more general.
The "kau" can express listener's knowledge.

Is it right?
I think it is very good definition of "kau".

ki'e
mi'e .iocikun.juj.

Jacob Errington

unread,
Oct 11, 2012, 11:16:27 PM10/11/12
to lojban-b...@googlegroups.com
coi

{kau} is only used inside abstractions to make indirect questions.

However, your intuition is right about placement of the particle. The cmavo you want to use for this is {xu}.

{.i do xu klama} "Is it you who goes?" (as opposed to someone else)
{.i do klama xu} "Are you *going*?" (as opposed to doing something else)

In both cases, the answer can be {na go'i} or {go'i}.

{.i do xu klama} "Is it you who goes?" {.i na go'i .i lo mamta ku klama} "No, my mother is going."
{.i do klama xu} "Are you *going*?" {.i na go'i .i mi ca citka} "No, I'm eating now."

.i mi'e la tsani mu'o


--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban-beginners/-/AtrR34fofOMJ.
To post to this group, send email to lojban-b...@googlegroups.com.
To unsubscribe from this group, send email to lojban-beginne...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban-beginners?hl=en.

Felipe Gonçalves Assis

unread,
Oct 11, 2012, 11:56:39 PM10/11/12
to lojban-b...@googlegroups.com
You appear to be describing a focus construction in Japanese.
English does it either by giving emphasis to the word "I" or "go"
(prosodic focus), or by a periphrastic construction such as
"It is I who goes" (syntactic focus).

CLL has a recently discussed example that suggests the use of
{kau} as a focus marker in lojban:
(8.4 in http://dag.github.com/cll/11/8). This use, however, is
disputed (see thread "[lojban] {kau} vs. {ba'e}" in the main
discussion list), the contrary opinion being that {ba'e} should
be used instead, mirroring the prosodic focus of English.

I have not participated in that discussion, but the fact is that the
use of {kau} as a focus marker is independent of its use as an
indirect question marker, since the preceding word is a question
word only in the latter. Also, the relation between emphasis and
focus is debatable.

I admit there is a certain ring to the use of {kau} as both indirect
question and focus marker: it is like the focus is replacing the
question word.

Anyway, to me, any informed use is valid, just keep in mind that
this is an undocumented part of lojban: {kau} has never been
clearly defined as focus marker, and {ba'e} is just vaguely defined
to correspond to emphasis.

mu'o
mi'e .asiz.

Felipe Gonçalves Assis

unread,
Oct 12, 2012, 12:13:17 AM10/12/12
to lojban-b...@googlegroups.com
On 12 October 2012 00:16, Jacob Errington <nict...@gmail.com> wrote:
> {.i do xu klama} "Is it you who goes?" (as opposed to someone else)
> {.i do klama xu} "Are you *going*?" (as opposed to doing something else)
>

I understood that "watashi ga/wa iku" are declarations, not questions...

In any case, it is relevant to point that, in the case of yes/no questions,
the non-initial position of {xu} marks the preceding element as a focus
of the question, no doubt or debate on that.

In other words, if {kau} marked the focus of sentences in general,
the following pairs could be considered equivalent.
{.i do xu klama} ~ {.i xu do kau klama}
{.i do klama xu} ~ {.i xu do klama kau}

mu'o
mi'e .asiz.

.iocikun.juj.

unread,
Oct 12, 2012, 1:10:04 AM10/12/12
to lojban-b...@googlegroups.com
ki'e la tsani
ki'e .asiz.

It's true that "watashi ga/wa iku" are declarations. These are not questions.

I understand that I can use {ba'e}.
{ba'e mi klama} = "watashi ga iku"
{mi ba'e klama} = "watashi wa iku"

But {ba'e} is vague.
To use kau may be more strict.

I think that listener's knowledge is essential to languages.

{mi djuno lodu'u la djan. kau klama} is "I know that it's John who go".
Listener may know someone goes, but may not know who goes.

It mean:
[Tom go, Jack go, John go ... ] -> [John go]
Before meaning list is narrowed to after meaning list.

I think its essential to language that [before meaning list] is narrowed to [after meaning list].
I want to strictly express this narrowing by {kau}.

I want to use {kau} outside {du'u}.
It's good that I can say following.

la djan. kau klama
(You know that someone goes. I inform you that it's John who goes)

And I can do ordering more strictly.

{mi viska da gi'e da prami ko} is "Please show me the person who loves you" or "Please be loved by the person who I see".
And {mi viska kau da gi'e da prami ko} is "Please show me the person who loves you".
[I don't see X and X loves you, I see X and X loves you ... ] -> [I see X and X loves you]
And {mi viska da gi'e da prami kau ko} is "Please be loved by the person who I see".
[I see X and X don't loves you, I see X and X loves you ...] -> [I see X and X loves you]

I want you to understand "before meaning list".
I don't know that this expression is appropriate.
And the idea is essential to languages, isn't it?

2012年10月12日金曜日 13時13分18秒 UTC+9 .asiz.:

.iocikun.juj.

unread,
Oct 12, 2012, 1:18:18 AM10/12/12
to lojban-b...@googlegroups.com

{mi viska da gi'e da prami ko} is "Please show me the person who loves you" or "Please be loved by the person who I see".
And {mi viska kau da gi'e da prami ko} is "Please show me the person who loves you".
[I don't see X and X loves you, I see X and X loves you ... ] -> [I see X and X loves you]
And {mi viska da gi'e da prami kau ko} is "Please be loved by the person who I see".
[I see X and X don't loves you, I see X and X loves you ...] -> [I see X and X loves you]


mi viska da gi'e da prami ko -> viska fa mi da gi'e prami fa da ko
mi viska kau da gi'e da prami ko -> viska kau fa mi da gi'e prami fa da ko
mi viska da gi'e da prami kau ko -> viska fa mi da gi'e prami kau fa da ko

gleki

unread,
Oct 12, 2012, 3:02:52 AM10/12/12
to lojban-b...@googlegroups.com
I see no problem here.

Japanese wa - ga are topic-comment markers.
Read the CLL on that. the CLL gives an example from Chinese that lacks separate words in this case for that.
But generally it's the same.
{zo'u} can do the trick.

.iocikun.juj.

unread,
Oct 12, 2012, 6:18:55 AM10/12/12
to lojban-b...@googlegroups.com
ki'e la gleki

I think some "wa" can be translate to use {kau} and some "wa" can be translate to use {zo'u}.

{zo'u} is great.
The Japanese sentence "zou wa hana ga nagai" is {lo xanto zo'u lo nazbi clani}.
It mean that noses of elephants are long.
To use {zo'u}, I can translate directly.

zou: elephant
hana: nose
nagai: long

mu'o

2012年10月12日金曜日 16時02分52秒 UTC+9 gleki:

mashers

unread,
Oct 12, 2012, 1:25:07 PM10/12/12
to lojban-b...@googlegroups.com
coi .iocikun.juj.

Is Japanese syntax generally constructed in this way? I find it fascinating that the emphasis you are making in the response is marked in the grammar of the sentence!

Felipe Gonçalves Assis

unread,
Oct 12, 2012, 2:07:11 PM10/12/12
to lojban-b...@googlegroups.com
On 12 October 2012 02:10, .iocikun.juj. <yoshiku...@gmail.com> wrote:
>
> I want to use {kau} outside {du'u}.
> It's good that I can say following.
>
> la djan. kau klama
> (You know that someone goes. I inform you that it's John who goes)
>

If {kau} is to be accepted as a focus marker inside {du'u}, it is only logical
that it is equally accepted in main bridi, yes.

> And I can do ordering more strictly.
>
> {mi viska da gi'e da prami ko} is "Please show me the person who loves you"
> or "Please be loved by the person who I see".
> And {mi viska kau da gi'e da prami ko} is "Please show me the person who
> loves you".
> [I don't see X and X loves you, I see X and X loves you ... ] -> [I see X
> and X loves you]
> And {mi viska da gi'e da prami kau ko} is "Please be loved by the person who
> I see".
> [I see X and X don't loves you, I see X and X loves you ...] -> [I see X and
> X loves you]
>

I think I understand where you are trying to get here, but I guess focus
is not the appropriate solution.

Focus is useful in imperatives just as in an other construction:
{ko ba'e jgari ta} / {ko jgari kau ta} -> "HOLD that" (Don't only touch it)
{ko jgari ba'e ta} / ko jgari ta kau} -> "Hold THAT" (Not something else)

I guess your sentence
> {mi viska da gi'e da prami ko} is "Please show me the person who loves you"
may be corrected as
{mi viska da .ije da prami ko}
rewritten as
{mi viska da poi [ke'a] prami ko}
or
{ko se prami da poi mi viska [ke'a]}

You correctly understand that, as it is, this request doesn't specify any
entity I am seeing that I want you to be loved by, or any entity that loves
you that I want you to show me. In fact, there may be neither, and you
may need to cure my blindness as well as gain the love of someone to
show me.

However, this is not a question of lack of focus, it is just a lack of
description:
{ko se prami lo se viska be mi}
"Be loved by the one I see."
vs.
{ko vi'argau mi lo prami be do}
"Show me the lover of yours"


> And the idea is essential to languages, isn't it?
>

I am no typologist, but I don't know of any language without focus.

On a theoretical level, we expect other's expressions to convey relevant
information, but sometimes, it is grammatically impossible to convey the
relevant information without putting in some more irrelevant facts, just
to build a grammatical phrase. Focus is then the way we can single out
the relevant among the by-products.

Pierre Abbat

unread,
Oct 12, 2012, 5:33:35 PM10/12/12
to lojban-b...@googlegroups.com
On Friday, October 12, 2012 03:18:55 .iocikun.juj. wrote:
> ki'e la gleki
>
> I think some "wa" can be translate to use {kau} and some "wa" can be
> translate to use {zo'u}.
>
> {zo'u} is great.
> The Japanese sentence "zou wa hana ga nagai" is {lo xanto zo'u lo nazbi
> clani}.
> It mean that noses of elephants are long.
> To use {zo'u}, I can translate directly.
>
> zou: elephant
> hana: nose
> nagai: long

I was just writing this exact example - one minor point, you need "cu" before
"clani" or "ku" after "nazbi".

Pierre
--
sei do'anai mi'a djuno puze'e noroi nalselganse srera

gleki

unread,
Oct 13, 2012, 4:04:02 AM10/13/12
to lojban-b...@googlegroups.com

In the parallel thread I completely forgot to explain why I thought {kau} could be used for focus of any sentence.

Look at the following example:
[.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu] .i la .djan. kau zekri prenu


Now imagine that we omit the sentences in brackets and let the context decide what's going on here.
Even then it's pretty simple to reconstruct first three sentences if only the last one is given.
But it just means that every time you use {kau} without any questions (even indirect questions), even then it doesn't mean that those questions are absent.

They still EXIST. They are just omitted by the speaker but can be easily resurrected from the sentences that are actually said.
Really it would be very verbous to say 
.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu.....................

It's just not worth it.


However, I doubt a bit that it is {kau}, not any other UI-cmavo that can serve this purpose.
Do we have a generalised UI-cmavo like we have {su'u} as a generalised NU-cmavo?
{ge'e} speaks about emotions  so it won't work.

Next, {paunai} is defined as a rhetorical question. I don't know exactly what "rhetorical" means. For me this glossword includes some emotions. But if {paunai} is just the opposite of {pau} and {pau} is just a question marker then for me {paunai=kau}.

Jacob Errington

unread,
Oct 13, 2012, 9:47:21 AM10/13/12
to lojban-b...@googlegroups.com
On 13 October 2012 01:04, gleki <gleki.is...@gmail.com> wrote:

In the parallel thread I completely forgot to explain why I thought {kau} could be used for focus of any sentence.

Look at the following example:
[.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu] .i la .djan. kau zekri prenu


Yeah, I don't know why you'd think the kau gets preserved. {makau} acts as one single unit, and gets replaced entirely when you want to insert the "answer" to the indirect question.
 

Now imagine that we omit the sentences in brackets and let the context decide what's going on here.
Even then it's pretty simple to reconstruct first three sentences if only the last one is given.
But it just means that every time you use {kau} without any questions (even indirect questions), even then it doesn't mean that those questions are absent.

They still EXIST. They are just omitted by the speaker but can be easily resurrected from the sentences that are actually said.
Really it would be very verbous to say 
.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu.....................

It's just not worth it.


Or we could just say {la djan zekri prenu}. In a real conversation, the difference between the two sentences, with and without kau, is purely textual, AFAICT. Using {kau} conveys no extra meaning, and therefore leads me to believe that it's pointless to use. 
 

However, I doubt a bit that it is {kau}, not any other UI-cmavo that can serve this purpose.
Do we have a generalised UI-cmavo like we have {su'u} as a generalised NU-cmavo?
{ge'e} speaks about emotions  so it won't work.


Guys, it's ba'e. Can we please stop trying to use kau for things other than indirect questions? :)
If you want to say "It's John, and not someone else, that is the criminal" then why are you all pining for {la djan kau co'e} rather than {ba'e la djan co'e} ?
 
Next, {paunai} is defined as a rhetorical question. I don't know exactly what "rhetorical" means. For me this glossword includes some emotions. But if {paunai} is just the opposite of {pau} and {pau} is just a question marker then for me {paunai=kau}.


A rhetorical question is a sentence that appears to be a question, but isn't actually asking anything. "Who could *do* such a thing!?" could be a rhetorical question. {kau} is not for "questions that aren't questions"; {kau} is for *indirect questions*. Those are both *different* things, and therefore have *different* cmavo, albeit both UI.

gleki

unread,
Oct 14, 2012, 7:31:06 AM10/14/12
to lojban-b...@googlegroups.com


On Saturday, October 13, 2012 5:47:42 PM UTC+4, tsani wrote:
On 13 October 2012 01:04, gleki <gleki.is...@gmail.com> wrote:

In the parallel thread I completely forgot to explain why I thought {kau} could be used for focus of any sentence.

Look at the following example:
[.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu] .i la .djan. kau zekri prenu


Yeah, I don't know why you'd think the kau gets preserved. {makau} acts as one single unit, and gets replaced entirely when you want to insert the "answer" to the indirect question.

This would make sense if {la djan. kau}  were absent in the CLL.
{kau} is glossed as "indirect question" and I can't see any questions in {la djan. kau}.

 

Now imagine that we omit the sentences in brackets and let the context decide what's going on here.
Even then it's pretty simple to reconstruct first three sentences if only the last one is given.
But it just means that every time you use {kau} without any questions (even indirect questions), even then it doesn't mean that those questions are absent.

They still EXIST. They are just omitted by the speaker but can be easily resurrected from the sentences that are actually said.
Really it would be very verbous to say 
.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu.....................

It's just not worth it.


Or we could just say {la djan zekri prenu}. In a real conversation, the difference between the two sentences, with and without kau, is purely textual, AFAICT. Using {kau} conveys no extra meaning, and therefore leads me to believe that it's pointless to use. 
 

However, I doubt a bit that it is {kau}, not any other UI-cmavo that can serve this purpose.
Do we have a generalised UI-cmavo like we have {su'u} as a generalised NU-cmavo?
{ge'e} speaks about emotions  so it won't work.


Guys, it's ba'e. Can we please stop trying to use kau for things other than indirect questions? :)

I'm not sure if emphasis and topic-comment structure are always identical.

Jacob Errington

unread,
Oct 14, 2012, 10:39:22 AM10/14/12
to lojban-b...@googlegroups.com
On 14 October 2012 07:31, gleki <gleki.is...@gmail.com> wrote:


On Saturday, October 13, 2012 5:47:42 PM UTC+4, tsani wrote:
On 13 October 2012 01:04, gleki <gleki.is...@gmail.com> wrote:

In the parallel thread I completely forgot to explain why I thought {kau} could be used for focus of any sentence.

Look at the following example:
[.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu] .i la .djan. kau zekri prenu


Yeah, I don't know why you'd think the kau gets preserved. {makau} acts as one single unit, and gets replaced entirely when you want to insert the "answer" to the indirect question.

This would make sense if {la djan. kau}  were absent in the CLL.
{kau} is glossed as "indirect question" and I can't see any questions in {la djan. kau}.


The CLL being wrong is new? :P
 
 

Now imagine that we omit the sentences in brackets and let the context decide what's going on here.
Even then it's pretty simple to reconstruct first three sentences if only the last one is given.
But it just means that every time you use {kau} without any questions (even indirect questions), even then it doesn't mean that those questions are absent.

They still EXIST. They are just omitted by the speaker but can be easily resurrected from the sentences that are actually said.
Really it would be very verbous to say 
.i paunai ma zekri prenu .i mi djuno lo du'u makau zekri prenu .i mi djuno lo du'u la .djan. kau zekri prenu.....................

It's just not worth it.


Or we could just say {la djan zekri prenu}. In a real conversation, the difference between the two sentences, with and without kau, is purely textual, AFAICT. Using {kau} conveys no extra meaning, and therefore leads me to believe that it's pointless to use. 
 

However, I doubt a bit that it is {kau}, not any other UI-cmavo that can serve this purpose.
Do we have a generalised UI-cmavo like we have {su'u} as a generalised NU-cmavo?
{ge'e} speaks about emotions  so it won't work.


Guys, it's ba'e. Can we please stop trying to use kau for things other than indirect questions? :)

I'm not sure if emphasis and topic-comment structure are always identical.
 

No, of course they aren't the same, and I didn't say they were. {zo'u} is for topic-comment structure, and {ba'e} is for emphasis. The CLL is right on that one.
 
If you want to say "It's John, and not someone else, that is the criminal" then why are you all pining for {la djan kau co'e} rather than {ba'e la djan co'e} ?
 
Next, {paunai} is defined as a rhetorical question. I don't know exactly what "rhetorical" means. For me this glossword includes some emotions. But if {paunai} is just the opposite of {pau} and {pau} is just a question marker then for me {paunai=kau}.


A rhetorical question is a sentence that appears to be a question, but isn't actually asking anything. "Who could *do* such a thing!?" could be a rhetorical question. {kau} is not for "questions that aren't questions"; {kau} is for *indirect questions*. Those are both *different* things, and therefore have *different* cmavo, albeit both UI.

.i mi'e la tsani mu'o

--
You received this message because you are subscribed to the Google Groups "Lojban Beginners" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban-beginners/-/76RjAIGKOLIJ.

la gleki

unread,
Dec 28, 2012, 8:01:47 AM12/28/12
to lojban-b...@googlegroups.com
Now I think that using {kau} as a topic marker will lead to some problems.

Topic is always in the left of the sentence. It's quite natural. At first you define the topic and then tell something about it in particular.
However, {kau} can be placed anywhere. Is it something that we want?
Reply all
Reply to author
Forward
0 new messages