lo selsanga pe la'o dy. Bodo Wartke .dy zi'e po'u la'o dy. Liebeslied .dy

41 views
Skip to first unread message

v4hn

unread,
Aug 14, 2012, 11:12:52 AM8/14/12
to loj...@googlegroups.com
coi rodo

ni'o la'o gy. Bodo Wartke .gy goi ko'a ge'u dotco larfi'i gi'e sanga
.i ko'a finti lo selsanga poi se cmene zoi Liebeslied zoi
to noi jibni se fanva fu lu lo prami selsanga li'u toi
.i ko'a troci lonu sanga lo pempau be sy be'o bau ro lo bangu pe le terdi
.i uasai ku'i mi pu facki lodu'u ko'a sanga py bau la .esperantos. joi
la .klingon. joi la .kvenjas. joi la .sindarin. ba'e joinai la .lojban.
.i mi djica lenu sabji pa lo lojbo pempau ku ko'a kei
.i mi troci finti ly. po'u di'e ge'u

.i mi djica lenu sanga fi do
bau ro lo bangu sepi'o ro lo zgitci
.i mi prami do

.i ta'o le glico pempau me di'e

zoi gy.
I want to sing in every language for you baby
and play on every instrument to say these words to you.
Believe me! It is true! I love you.
.gy

.i mi pu rejgau fi lonu mi sanga lo mi lojbo pempau joi lo glico pempau joi lo dotco pempau
fe lo seljudri be zoi gy. http://v4hn.de/files/liebeslied-jbo-en-de.mp3 .gy
.i ko'a sabji ro lo ko'a pempau noi se judri zoi gy. http://www.bodowartke.de/liebesliedgenerator/llg_flash_2012.php .gy

.i ma pinka fi do

mu'o mi'e la .van.

iesk

unread,
Aug 14, 2012, 4:22:19 PM8/14/12
to loj...@googlegroups.com, m...@v4hn.de


coi la .van.

.i ckire do lonu jungau .i la'o dy. Liebeslied .dy pe la .bodos.vartkys. mi slabu .i puku mi facki lo du'u da ja'a fanva fi la .lojban. .i xe fanva

«lu tu'e
 .ai mi dei do ge sanga bau ro bangu
 gi se pi'o ro cabra ku zgicusku
 .i .iu ju'o
 mi prami do
tu'u li'u»

(to .i .e'u co'e la'o zoi. https://groups.google.com/forum/?fromgroups#!topic/lojban/PCfF5t3ydCw[1-25] .zoi toi)

.i lo pe ti'e la .aleks. .kox. xe fanva cu xauzma ba'e pe'i lo do .io xe fanva fo lo ni zgike mapti kei .e lo ni bangu melbi

.i ku'i mi ja'a ckire do lo .io nu fanva .i jinvi lo du'u lo du'u sei gerna .u'i se'u so'o xe fanva cu zasti kei cu cinri je se kucli je xe cilre

.i .ui nai la .bodos. na sanga vau za'a

mu'o mi'e .iesk.

v4hn

unread,
Aug 16, 2012, 7:57:40 AM8/16/12
to loj...@googlegroups.com
On Tue, Aug 14, 2012 at 01:22:19PM -0700, iesk wrote:
> coi la .van.
>
> .i ckire do lonu jungau .i la'o dy. Liebeslied .dy pe la .bodos.vartkys. mi
> slabu .i puku mi facki lo du'u da ja'a fanva fi la .lojban.

.iua mi na pu djuno

> .i xe fanva
>
> «lu tu'e
> .ai mi dei do ge sanga bau ro bangu
> gi se pi'o ro cabra ku zgicusku
> .i .iu ju'o
> mi prami do
> tu'u li'u»

.iuisai mi nelci ti .i lo mi xe fanva be ly ku pare'u jai se troci be mi

> .i jinvi lo du'u lo du'u sei gerna
> .u'i se'u so'o xe fanva cu zasti kei cu cinri je se kucli je xe cilre

je'e

> .i .ui nai la .bodos. na sanga vau za'a

.i xu la .bodos. pu spuda lo mrilu gi'e cusku lo krinu be lonu na pilno lo selsabji
.i da'i ganai mi ba penmi by gi mi te preti fo by

Paul Predkiewicz

unread,
Aug 16, 2012, 8:53:22 AM8/16/12
to loj...@googlegroups.com
Sorry to interrupt in english, but my lojban is only as good as to understand this is about the song Liebeslied from Bodo Wartke and that you have come up with a verse already, but cant find the gismu rimni or the rafsi rim anywhere here. I just wanted to say that bodo has also provided a blueprint for the rhyming.

i.e.:
Ich will's in allen Sprachen für dich singen(a),
auf allen Instrumenten zum Erklingen(a) bringen(a).
Ich(b)   [kinda long pause]
liebe dich(b).

in which (a) is the first rhymeword,
(a)(a) two words which rhyme on the first rhymeword,
(b) is the second rhymeword and again
(b) a word that rhymes on the second rhymeword.

in case you got all that already covered, disregard all of this

mlatu

2012/8/16 v4hn <m...@v4hn.de>

v4hn

unread,
Aug 16, 2012, 11:05:19 AM8/16/12
to loj...@googlegroups.com
coi do

On Thu, Aug 16, 2012 at 02:53:22PM +0200, Paul Predkiewicz wrote:
> Sorry to interrupt in english, but my lojban is only as good as to
> understand this is about the song Liebeslied from Bodo Wartke and that you
> have come up with a verse already, [...]

No problem at all, I started to learn Lojban only recently.
It takes me a lot of time to write mails in lojban.

> but cant find the gismu rimni or the
> rafsi rim anywhere here. I just wanted to say that bodo has also provided a
> blueprint for the rhyming.
>
> i.e.:
> Ich will's in allen Sprachen für dich singen(a),
> auf allen Instrumenten zum Erklingen(a) bringen(a).
> Ich(b) [kinda long pause]
> liebe dich(b).
>
> in which (a) is the first rhymeword,
> (a)(a) two words which rhyme on the first rhymeword,
> (b) is the second rhymeword and again
> (b) a word that rhymes on the second rhymeword.

He also mentions that variations from this scheme are possible.
I wrote earlier that my translation was only a first attempt,
but iesk found this older translation {ti'e} by Alex Koch
and this one is quite nice, though I think it's a little bit lengthy.


v4hn

Paul Predkiewicz

unread,
Aug 16, 2012, 11:18:12 AM8/16/12
to loj...@googlegroups.com
Well i think if you could omit the {lu tu'e} in the beginning and the {tu'u li'u} in the end, it might work quite well.
though the second line needs to contain another two or three silable word, or else the timing won't fit so nicely...
best would be one ending in an "gu" or "ku"

2012/8/16 v4hn <m...@v4hn.de>

iesk

unread,
Aug 18, 2012, 9:30:09 AM8/18/12
to loj...@googlegroups.com
Le jeudi 16 août 2012 17:18:12 UTC+2, kamisori a écrit :
Well i think if you could omit the {lu tu'e} in the beginning and the {tu'u li'u} in the end, it might work quite well.
.i te pandi co bacru .i na pagbu lo se sanga
That’s spoken punctuation. It doesn’t belong to the lyrics.
Das ist verbale Zeichensetzung und gehört nicht zum Liedtext.

http://dag.github.com/cll/19/2/
http://dag.github.com/cll/19/9/

 
though the second line needs to contain another two or three silable word, or else the timing won't fit so nicely...
best would be one ending in an "gu" or "ku"

.i pe'i mapti lo rilti

ai mị dei do ge sạnga bau ro bạngu
gi sẹ pi'ọ ro cạbra ku zgicụsku
.i .iụ ju'ọ
mi prạmi do

.i .ie ca'i la .bodos. tu'a so'o lo pemci tarmi cu se curmi .i ja'o loi lojbo ka'e finti so'i lo drata xe fanva .i ku'i mi na djuno lodu'u .uinai mu'i makau la .bodos. genai za'a jmina lo lojbo mela.libyslit. ginai ti'e spuda
Yes, according to Bodo several variations are possible. So, Lojbanists can make different translations. I don’t know how Bodo didn’t add a Lojban version and didn’t respond.
Bodo sagt ja, daß mehrere Versformen möglich sind. Die Lojbanisten könnten also noch weitere Übersetzungen fertigen. Ich weiß aber nicht, warum Bodo keine Lojban-Version hinzugefügt hat und warum er anscheinend auch nicht geantwortet hat.

.i la'apei by. vy. za'o sanga la .libyslit. vau .u'i ru'e

iesk

Michael Turniansky

unread,
Sep 14, 2012, 3:32:04 PM9/14/12
to loj...@googlegroups.com
  Just a minor grammatical point about your letter, van..  You cannot cluster a cmavo together with another if it starts with a vowel (as you have done below with *"i.iua" and *".iuisai" without putting a pause (period or space in written lojban) before it.  So that should be either ".i uisai" or ".i.uisai"

      --gejyspa

selpa'i

unread,
Sep 14, 2012, 3:50:07 PM9/14/12
to loj...@googlegroups.com
Am 14.09.2012 21:32, schrieb Michael Turniansky:
> Just a minor grammatical point about your letter, van.. You cannot
> cluster a cmavo together with another if it starts with a vowel (as
> you have done below with *"i.iua" and *".iuisai" without putting a
> pause (period or space in written lojban) before it. So that should be
> either ".i uisai" or ".i.uisai"

Without wanting to re-open this debate from a while ago, I still don't
like seeing this kind of information being spread. Neither a pause nor a
period nor a space is needed in written or spoken Lojban. ".iua"
unambiguously fragments into .i ua. I'd repeat myself and say that ua is
a CV, but I don't want to get falsely flamed again for saying the truth.
For completeness sake however, I should add that there are no words that
begin with a (phonetic) vowel in Lojban, so the rule you formulated
above just never applies. What I'm saying might seem wrong or alien to
you, but I've done enough research and have thought about this topic at
length. You are free to reject it, of course. If you'd like to discuss
this in more depth (since it probably makes no sense to you here without
the proper context), feel free to open a new thread.
I'd like to mention that because you are by far not the only one who
gets confused about Lojban's "denpa bu", I created a slightly altered
way of writing lojban some time ago which makes it impossible to
mispronounce things and which always clearly indicates what's going on
phonologically/morphologically (the goal being actual full isomorphism
within the limits of a simple latin alphabet using only a few
diacritics). Some people like it, some don't. Some like it but don't use
it, others actually have adopted it. It does help people though to not
be confused about Lojban's phonology.
Bottomline, you really don't need a "denpa bu" in the sentence in question.

mu'o mi'e la selpa'i

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

doị mèlbi mlenì'u
.i do càtlu ki'u
ma fe la xàmpre ŭu
.i do tìnsa càrmi
gi'e sìrji se tàrmi
.i taị bo pu cìtka lo gràna ku

Michael Turniansky

unread,
Sep 14, 2012, 4:52:44 PM9/14/12
to loj...@googlegroups.com
  Well, of course, I haven't seen the debate yet (still being more than 3 weeks behind on email) and therefore don't know if you've already addressed the following question, but EVEN IF I accept arguendo your statement that "ua" is a CV (although I don't), then you must ALSO assert that "iu" is a CV. So how do you distinguish between "iu isai..." and  "i ui sai..." if all the person put down (as here) was "iuisai..."? (And let's not forget that plenty of combinations like "iau" are valid in names and fu'ivla and are different than "i au and ia u"  The pauses need to be present to distinguish, or there has to be an unambiguous rule of decomposition, which I don't believe there can be, but I will wait further commenting upon my reading of what youve already written onthe topic.

               --gejyspa



--
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+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.


v4hn

unread,
Sep 14, 2012, 4:59:50 PM9/14/12
to loj...@googlegroups.com
On Fri, Sep 14, 2012 at 09:50:07PM +0200, selpa'i wrote:
> Am 14.09.2012 21:32, schrieb Michael Turniansky:
> >Just a minor grammatical point about your letter, van.. You cannot
> >cluster a cmavo together with another if it starts with a vowel
> >(as you have done below with *"i.iua" and *".iuisai" without
> >putting a pause (period or space in written lojban) before it. So
> >that should be either ".i uisai" or ".i.uisai"

Thank you for your comment Michael, I actually agree with you
in that it's hard to read text without these spaces / pauses.

{doi la selpa'i} I think I understand your point.
Do you at least agree, that _it makes sense_ to insert spaces or pauses
in such places with the "standard" alphabet because of garden path effects?
I mean if you mistake { ... .iuaiiue ... } for { ... .iuaiiu ... }
when chatting with your partner that could have a serious impact
on your relationship...


v4hn

selpa'i

unread,
Sep 14, 2012, 5:11:51 PM9/14/12
to loj...@googlegroups.com
Am 14.09.2012 22:52, schrieb Michael Turniansky:
> Well, of course, I haven't seen the debate yet (still being more
> than 3 weeks behind on email) and therefore don't know if you've
> already addressed the following question,

It happened long ago, I can't remember when exactly, but it's been
almost a year, but at least half a year.

> but EVEN IF I accept arguendo your statement that "ua" is a CV
> (although I don't), then you must ALSO assert that "iu" is a CV.

Yes, iu is CV (bearing though that this C is not the same type of C that
is used to create brivla clusters under the current rules. changing this
would in theory be possible however, which would cause things like bioka
to be CCVCV brivla. not currently the case, however.)

> So how do you distinguish between "iu isai..." and "i ui sai..." if
> all the person put down (as here) was "iuisai..."?

The problem here is because it is technically accepted to elide denpa bu
when a space is present, people aren't sure when denpa bu is actually
necessary and when it isn't. It also makes people confused about what
the point of the letter is. I encourage to always include all denpa bu
that are necessary and to elide all that can be elided if one so wishes.

If you write down "iu isai" then that becomes {iu .i sai}. If you write
it down without spaces, which is the more interesting case and more
closely resembles the actual speechstream, then you have to make sure
the automatic initial denpa bu that people are using does what you want
it to do. {iuisai} would become {.i ui sai} if you allow denpa bu to be
elided. If not, it would be ungrammatical (because after iu, the next i
is not a lojban word and it fails to parse).

> (And let's not forget that plenty of combinations like "iau" are valid
> in names and fu'ivla and are different than "i au and ia u" The
> pauses need to be present to distinguish, or there has to be an
> unambiguous rule of decomposition, which I don't believe there can be,
> but I will wait further commenting upon my reading of what youve
> already written onthe topic.

There is indeed such a rule of decomposition and only because it exists
can I in good conscience preach this. You just need to be aware that the
semi-vowels are not vowels, they are [w] and [j] in IPA respectively. If
you consider speechstreams, then those are distinct from the actual
Lojban vowels (V). "iua" is only possible in cmevla and zi'evla. A cmavo
can't have three VVV in a row, so the cmavo clusters always decompose
unambiguously. If I didn't answer all your questions, please feel free
to through more questions or tests at me.

selpa'i

unread,
Sep 14, 2012, 5:11:59 PM9/14/12
to loj...@googlegroups.com
Am 14.09.2012 22:59, schrieb v4hn:
> On Fri, Sep 14, 2012 at 09:50:07PM +0200, selpa'i wrote:
>> Am 14.09.2012 21:32, schrieb Michael Turniansky:
>>> Just a minor grammatical point about your letter, van.. You cannot
>>> cluster a cmavo together with another if it starts with a vowel
>>> (as you have done below with *"i.iua" and *".iuisai" without
>>> putting a pause (period or space in written lojban) before it. So
>>> that should be either ".i uisai" or ".i.uisai"
> Thank you for your comment Michael, I actually agree with you
> in that it's hard to read text without these spaces / pauses.
>
> {doi la selpa'i} I think I understand your point.
> Do you at least agree, that _it makes sense_ to insert spaces or pauses
> in such places with the "standard" alphabet because of garden path effects?

It doesn't make sense to add denpa bu, but you are free to add spaces to
facilitate reading. The space is entirely optional though.


> I mean if you mistake { ... .iuaiiue ... } for { ... .iuaiiu ... }
> when chatting with your partner that could have a serious impact
> on your relationship...

Can't happen. It's clearly {.i ua ii ue}. The second one is not grammatical.

Jorge Llambías

unread,
Sep 14, 2012, 5:36:26 PM9/14/12
to loj...@googlegroups.com
On Fri, Sep 14, 2012 at 6:11 PM, selpa'i <sel...@gmx.de> wrote:
>
> A cmavo can't have
> three VVV in a row, so the cmavo clusters always decompose unambiguously.

I completely agree with selpa'i, but that bit can be confusing. "iai",
"iau", "iei", "ioi", "uai", "uau", "uei", "uoi" should all be valid
cmavo. I'm not sure if you are counting the first letter in those as a
V.

mu'o mi'e xorxes

selpa'i

unread,
Sep 14, 2012, 5:47:41 PM9/14/12
to loj...@googlegroups.com
I think this is untrodden territory which is open to future changes,
possibly. My considerations so far included mostly words that already
are part of the grammar. I agree that those words could in theory
constitute valid cmavo. The first letter in all of them is a C in my
opinion. I'm not sure what the consequences would be of allowing these
cmavo into the language, if any. I wouldn't be opposed to it though, I
think. Maybe when we run out of cmavo space, we can come back and scoop
up these additional CVV.

Michael Turniansky

unread,
Sep 15, 2012, 11:56:39 PM9/15/12
to loj...@googlegroups.com
On Fri, Sep 14, 2012 at 5:11 PM, selpa'i <sel...@gmx.de> wrote:
Am 14.09.2012 22:52, schrieb Michael Turniansky:

  Well, of course, I haven't seen the debate yet (still being more than 3 weeks behind on email) and therefore don't know if you've already addressed the following question,

It happened long ago, I can't remember when exactly, but it's been almost a year, but at least half a year.


but EVEN IF I accept arguendo your statement that "ua" is a CV (although I don't), then you must ALSO assert that "iu" is a CV.

Yes, iu is CV (bearing though that this C is not the same type of C that is used to create brivla clusters under the current rules. changing this would in theory be possible however, which would cause things like bioka to be CCVCV brivla. not currently the case, however.)


So how do you distinguish between "iu isai..." and  "i ui sai..." if all the person put down (as here) was "iuisai..."?

The problem here is because it is technically accepted to elide denpa bu when a space is present, people aren't sure when denpa bu is actually necessary and when it isn't. It also makes people confused about what the point of the letter is. I encourage to always include all denpa bu that are necessary and to elide all that can be elided if one so wishes. 

If you write down "iu isai" then that becomes {iu .i sai}. If you write it down without spaces, which is the more interesting case and more closely resembles the actual speechstream, then you have to make sure the automatic initial denpa bu that people are using does what you want it to do. {iuisai} would become {.i ui sai} if you allow denpa bu to be elided. If not, it would be ungrammatical (because after iu, the next i is not a lojban word and it fails to parse).

 
  What are you talking about?  The next i IS a lojbanic word, and parses fine (If you are asserting that a sentence may not be simply an attitudinal, that's not true.  Or perhaps you are asserting that sai may not stand alone, which is also untrue).  And I never suggested he couldn't elide the denpa bu.  I explicitly said:

.i uisai" or ".i.uisai"
 
  I don't use usually use denpa bu except before "i" or in educational settings.


(And let's not forget that plenty of combinations like "iau" are valid in names and fu'ivla and are different than "i au and ia u"  The pauses need to be present to distinguish, or there has to be an unambiguous rule of decomposition, which I don't believe there can be, but I will wait further commenting upon my reading of what youve already written onthe topic.

There is indeed such a rule of decomposition and only because it exists can I in good conscience preach this. You just need to be aware that the semi-vowels are not vowels, they are [w] and [j] in IPA respectively. If you consider speechstreams, then those are distinct from the actual Lojban vowels (V). "iua" is only possible in cmevla and zi'evla. A cmavo can't have three VVV in a row, so the cmavo clusters always decompose unambiguously. If I didn't answer all your questions, please feel free to through more questions or tests at me.

   Yes, I have no misunderstanding about [w] and [j].  But they are a natural artifact of the pronouncing of two vowels close together.  For example, if you (as some do, but I don't) say that the name le jegvo cevni is la'o jegvo iaue jegvo, you can pronounce it as a single vocalic "flow" from one to the other, and they naturally cause a [w] and [j] (and an [aʊ̯]) to appear (Go ahead, I'll wait (*hums*))  That's all the morphologic rule means, and why it's necessary to separate out the vowel pairs that should NOT be pronounced as diphthongs or semivowels+vowel.  Nonetheless, I will agree that in speech, it's not necessary to pause between the words, because the pronunciation as a semivowel and not a vowel will naturally cause the break, but in print (which is what we were addressing here, in case you've forgotten), it IS, because otherwise the decomposition is ambiguous.

            --gejyspa

selpa'i

unread,
Sep 16, 2012, 8:32:59 AM9/16/12
to loj...@googlegroups.com
Am 16.09.2012 05:56, schrieb Michael Turniansky:

If you write down "iu isai" then that becomes {iu .i sai}. If you write it down without spaces, which is the more interesting case and more closely resembles the actual speechstream, then you have to make sure the automatic initial denpa bu that people are using does what you want it to do. {iuisai} would become {.i ui sai} if you allow denpa bu to be elided. If not, it would be ungrammatical (because after iu, the next i is not a lojban word and it fails to parse).

 
  What are you talking about?  The next i IS a lojbanic word, and parses fine (If you are asserting that a sentence may not be simply an attitudinal, that's not true.  Or perhaps you are asserting that sai may not stand alone, which is also untrue). 

No, I don't assert either of those things. Please re-read what I said.


And I never suggested he couldn't elide the denpa bu.  I explicitly said:

.i uisai" or ".i.uisai"

You wrote:

So how do you distinguish between "iu isai..." and  "i ui sai..." if all the person put down (as here) was "iuisai..."?

And I answered that question, once with and once without denpa bu elision being allowed. If denpa bu cannot be elided, then "iuisai" must become *{iu i sai}, which is not grammatical, because just "i" [i] is not a Lojban word. {.i} [ʔi] would be the Lojban word, but "i" is not. A Lojban word cannot begin with a vowel.


 
  I don't use usually use denpa bu except before "i" or in educational settings.

Do you use it or not? You can do what you want, I just recommend always including all necessary denpa bu because it causes less confusion about the rules and is a better representation of what is going on morphologically.




(And let's not forget that plenty of combinations like "iau" are valid in names and fu'ivla and are different than "i au and ia u"  The pauses need to be present to distinguish, or there has to be an unambiguous rule of decomposition, which I don't believe there can be, but I will wait further commenting upon my reading of what youve already written onthe topic.

There is indeed such a rule of decomposition and only because it exists can I in good conscience preach this. You just need to be aware that the semi-vowels are not vowels, they are [w] and [j] in IPA respectively. If you consider speechstreams, then those are distinct from the actual Lojban vowels (V). "iua" is only possible in cmevla and zi'evla. A cmavo can't have three VVV in a row, so the cmavo clusters always decompose unambiguously. If I didn't answer all your questions, please feel free to through more questions or tests at me.

   Yes, I have no misunderstanding about [w] and [j].  But they are a natural artifact of the pronouncing of two vowels close together.  For example, if you (as some do, but I don't) say that the name le jegvo cevni is la'o jegvo iaue jegvo, you can pronounce it as a single vocalic "flow" from one to the other, and they naturally cause a [w] and [j] (and an [aʊ̯]) to appear (Go ahead, I'll wait (*hums*)) 

I would say for example:
la. iaues. cu cevni lo jegvo
[laʔ 'ja.wɛsʔ ʃu 'ʃɛvni lo ʒɛgvo]

iaues can only be ['ja.wɛs] in Lojban, otherwise you'd end up with a syllable without onset */es/ and Lojban doesn't like that. That's why it inserts denpa bu (which is a consonant) in otherwise empty onsets, just like some natlangs do (e.g. German).

[w] and [j] are not just natural artifacts, as you say. They have distinct properties in the morphology in that they don't act as vowels at all. This is not just a case of different ways of realizing a set of phonemes.


That's all the morphologic rule means, and why it's necessary to separate out the vowel pairs that should NOT be pronounced as diphthongs or semivowels+vowel.  Nonetheless, I will agree that in speech, it's not necessary to pause between the words, because the pronunciation as a semivowel and not a vowel will naturally cause the break, but in print (which is what we were addressing here, in case you've forgotten), it IS, because otherwise the decomposition is ambiguous.

The idea is that speech and writing are the same thing, that's one of Lojban's selling points, and, to a large degree, it works even without modifications to the script. The decomposition is not ambiguous. The string of cmavo gets decomposed from left to right, and if you reach a word that can't exist (like "i"), you know it's not a grammatical text. I think I have shown how the decomposition works in the string in question.

Michael Turniansky

unread,
Sep 16, 2012, 11:08:31 AM9/16/12
to loj...@googlegroups.com
  Ah, so NOW I see where you are coming from.  You believe that, contrary to the CLL, "a word must never begin with a vowel", and that the denpa bu must always be a glottal stop, not just a pause.  Both are contradicted by Chapter 3 of the CLL, and lojban as it is actually used.  Since are talking about some theoretical language that you use, and not lojban, there is no point in continuing this discussion, as we are talking about two different languages. 
             --gejyspa


--
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.

selpa'i

unread,
Sep 16, 2012, 11:28:05 AM9/16/12
to loj...@googlegroups.com
Am 16.09.2012 17:08, schrieb Michael Turniansky:
> Ah, so NOW I see where you are coming from. You believe that,
> contrary to the CLL, "a word must never begin with a vowel", and that
> the denpa bu must always be a glottal stop, not just a pause. Both
> are contradicted by Chapter 3 of the CLL,

The CLL's view on this topic is not very coherent. The CLL simply
overlooks a lot of this and is therefore outdated.

> and lojban as it is actually used.

As it is used by some. Those that understand the things I'm trying to
explain here (and agree with it) have already changed their way of using
Lojban to make it make more sense. You are not aware of all Lojban usage
out there, so this is an argument out of ignorance.

> Since are talking about some theoretical language that you use, and
> not lojban, there is no point in continuing this discussion, as we are
> talking about two different languages.

If by Lojban you mean what the CLL describes, then yes, that is not a
language I'm speaking and nor is anybody else I consider a good speaker
of Lojban. There is no point in bringing up the CLL in matters that it
clearly is either inconsistent about or that it adresses poorly. I'm
sure you are aware of at least some changes that have been made to the
original Language described in the CLL. Do you object to all of them
because the CLL contradicts them, or do you agree with them because they
actually make a lot of sense?

Michael Turniansky

unread,
Oct 10, 2012, 2:54:00 PM10/10/12
to loj...@googlegroups.com
On Sun, Sep 16, 2012 at 11:28 AM, selpa'i <sel...@gmx.de> wrote:
Am 16.09.2012 17:08, schrieb Michael Turniansky:

  Ah, so NOW I see where you are coming from.  You believe that, contrary to the CLL, "a word must never begin with a vowel", and that the denpa bu must always be a glottal stop, not just a pause.  Both are contradicted by Chapter 3 of the CLL,

The CLL's view on this topic is not very coherent. The CLL simply overlooks a lot of this and is therefore outdated.

  I find no incoherence on the topic nor outdatedness.  YMMV 


and lojban as it is actually used.

As it is used by some. Those that understand the things I'm trying to explain here (and agree with it) have already changed their way of using Lojban to make it make more sense. You are not aware of all Lojban usage out there, so this is an argument out of ignorance.

   No, but I have heard a lot of lojban over the past 8 years.  And with the exception of one person (and I guess, you), I haven't heard anyone treat them as glottal stops.
 

Since are talking about some theoretical language that you use, and not lojban, there is no point in continuing this discussion, as we are talking about two different languages.

If by Lojban you mean what the CLL describes, then yes, that is not a language I'm speaking and nor is anybody else I consider a good speaker of Lojban. There is no point in bringing up the CLL in matters that it clearly is either inconsistent about or that it adresses poorly. I'm sure you are aware of at least some changes that have been made to the original Language described in the CLL. Do you object to all of them because the CLL contradicts them, or do you agree with them because they actually make a lot of sense?

  Well, then, I guess you don't consider me a good speaker of lojban, since I follow the CLL, except perhaps in the one or two areas over the past decades where the BPFK has actually made a decision contrary to it (but  AFAIK nothing has been finalized by them in many many years).   But I don't see that in THIS matter I don't see any contradictory ruling by "the powers that be" And while CLL may or may not be the ultimate authority on "lojban as she is spoke", it's the best and most complete one out there to date.  Feel free to write another if you feel it's inadequate.  But I like to have my bible.
      --gejyspa

Reply all
Reply to author
Forward
0 new messages