Proposal for Katakana Lojban orthography.

[Jonathan Jones]

I've secretly wanted to do this for a long time, and last night I finally got around to it. The following is my proposal for a Lojban orthography using Katakana. The example is the {le fasnu} example used to show the various Lojban orthographies at http://arj.nvg.org/lojban/orthography.html

(Note, you need to have Japanese fonts installed to see it all correctly. The specific font name is "SimSun".)

Proposal for Katakana orthography

In keeping with the practice in nihongo of having no spaces, all {denpabu} are required. This also means one is required immediately prior to a selbri.

With a few exceptions, katakana are used as normal, so for example kakne is カクネ.

All 'u' kana are pronounced without the 'u' sound, unless followed by ｳ, as in nu: ヌウ.

Any 'CuV' is pronounced as 'CV', as in denpa: デンパ or ヅエンプア.

ｼ is pronounced {cy.}, and followed by a vowel or 'y' kana, as in cusku: シウスクウ or シュスクウ.
To write {si}, use スイ.

ジ is pronounced {jy.} and follows the same rules as シ, as in djica: ヅジイシャ.
To write {zi}, use ズイ.

フ is {fy.}, and follows the same rules as シ, as in frili: フリリ″.
To write {.y'u}, use ハウ.

ワ is pronounced {vy.}, and followed by a vowel, as in vu: ワウ.

The 'r' kana with “ is pronounced 'l', as in lo: ラ″.

The 'k' kana with º is pronounced {xy.} as in xrula: クºルウラ″ア.

ヲis pronounced {.y.}

An example:

レ″フアスヌウ

レ″フオンカºシュ。サンガ。イラ″。アリ″ス。ゴイコハフラチラ″ヘヂハウレ″ヌウシパシュレ″フオンカº。イコハケルロ″ガンセル″ショイ。アリ″ス。イミヅラ″。リク。イアホドカンロリ″ハウ。イル″ウショイドイ。リク。イ。イアミカンロ。イアホドゴヒリ″ウ。イル″。イア。イミヅジイシャロ″ヌウヅジウノレ″ヌドバカンサミレ″グウスタケワアンシイサンミシャラ″。ゼヅジエヅ。ポイバル″ワイジイブニ。イアホミホテチグニリ″ハウ

。イコハスプダゾヘル″コデンパ。イミニツシュレ″ヌシャツル″レ″ミデツリリ″ステ。ヲ。ヲ。ヲ。イミズイフレ″。イ。アウミカンサ。イマツシイカリ″ハウ

。イラ″。リク。ゴイコヘスプダゾヘル″ミバシペヅロ″グウスタジウブメシャラ″。カºシャシ。イ。アウドヅジイシャシュクºナダポイフアスヌウ。イペイスキナ。イペイズギケチグニ。イペイカºジミチグニ。イペイヅラシイ。イペイダンスリ″ウ

。イラ″。アリ″ス。ダンフウル″ミネル″シイロ″イズギケヅラシイ。イ。イハ。アウラ″レ″ミゼﾗﾌﾞﾙ″。セチグニシャタ。イミネル″シロ″イヌダンスウリ″ウ

I appreciate any and all critiques and suggestions regarding this proposal.

--
mu'o mi'e .aionys.

.i.a'o.e'e ko klama le bende pe denpa bu

[tijlan]

Some Japanese jbopre have attempted the Katakana transliteration for gismu and cmavo on the jbo-pon dictionary (http://www.editgrid.com/explore/user/tijlan/%E3%83%AD%E3%82%B8%E3%83%90%E3%83%B3%E8%BE%9E%E6%9B%B8) which i host on my EditGrid account.

I myself am a Japanese speaker and have a good command of its orthography. Katakana, like Hiragana, is a syllabary, meaning that it's not cut out for representing single consonants. Every brivla has a consonant cluster, and no Katakana except { ン } [n] can adequately represent that. Cmavo, on the other hand, have no such cluster and can at least by principle be accommodated by a syllbary. More suited for brivla -- content words -- is Kanji, the logograms. For example:

mi carna
ミ　回

ta tricu cmana
タ　木　山

Logograms, when you know them, are a great potential for speed reading.

Katakana, however, has other difficulties in substituting for ASCII. It doesn't distinguish /r/ and /l/, for instance. As you've shown above, { ri } and { li } both become { リ }. { 'a } and { xa } become { ハ }. And so on. One workaround may be to use Hiragana for one of each pair. Say, { ri | り } (Hiragana) & { li | リ } (Katakana), and { 'a | は } (Hiragana} & { xa | ハ } (Katakana}. And certain syllables would have to be represented with two letters, such as { cu | シュ }, which is a combination of { シ } and { ュ }, which is the smaller of { ユ }. It can get quite complicated for non-natives.

moklu gi'e pimlu
口　ギへ　羽

finpe vi lo rirxe midju
魚　ヴィ　ろ　川　中

[Jonathan Jones]

2010/9/2 tijlan <jbot...@gmail.com>

Some Japanese jbopre have attempted the Katakana transliteration for gismu and cmavo on the jbo-pon dictionary (http://www.editgrid.com/explore/user/tijlan/%E3%83%AD%E3%82%B8%E3%83%90%E3%83%B3%E8%BE%9E%E6%9B%B8) which i host on my EditGrid account.

I myself am a Japanese speaker and have a good command of its orthography. Katakana, like Hiragana, is a syllabary, meaning that it's not cut out for representing single consonants. Every brivla has a consonant cluster, and no Katakana except { ン } [n] can adequately represent that. Cmavo, on the other hand, have no such cluster and can at least by principle be accommodated by a syllbary. More suited for brivla -- content words -- is Kanji, the logograms. For example:

mi carna
ミ　回

ミシャルナ

ta tricu cmana
タ　木　山

タツリシュシマナ

Logograms, when you know them, are a great potential for speed reading.

Katakana, however, has other difficulties in substituting for ASCII. It doesn't distinguish /r/ and /l/, for instance. As you've shown above, { ri } and { li } both become { リ }. { 'a } and { xa } become { ハ }. And so on. One workaround may be to use Hiragana for one of each pair. Say, { ri | り } (Hiragana) & { li | リ } (Katakana), and { 'a | は } (Hiragana} & { xa | ハ } (Katakana}. And certain syllables would have to be represented with two letters, such as { cu | シュ }, which is a combination of { シ } and { ュ }, which is the smaller of { ユ }. It can get quite complicated for non-natives.

I addressed these issues in the proposal. I'm going to assume since you b=mentioned these issues and did not remark on my solution that you didn't actually read my proposal.

moklu gi'e pimlu
口　ギへ　羽

モクル″ギヘピムル″

finpe vi lo rirxe midju
魚　ヴィ　ろ　川　中

フインペワイロ″リルケºミヅジウ

Admittedly, not being a Japanese speaker, I wouldn't be able to assign kanji to all the jbovalsi. I chose katakana because it is used for foreign words, and while it is a syllabary, not an alphabet, with a little wrangling it can be used.

--

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.

[tijlan]

On 2 September 2010 20:00, Jonathan Jones <eye...@gmail.com> wrote:

ｼ is pronounced {cy.}, and followed by a vowel or 'y' kana, as in cusku: シウスクウ or シュスクウ.

The latter one should be シュスク (without ウ at the end).

To write {si}, use スイ.

To write {zi}, use ズイ.

スイ is more like {sui}. I suggest スィ. Likewise, ズィ for {zi}.

ジ is pronounced {jy.} and follows the same rules as シ, as in djica: ヅジイシャ.

or ヂシャ. {tci} can unambiguously be チ, and {dji} ヂ.

ヲis pronounced {.y.}

To write {y'u}, use ハウ.

ハウ is most probably {xau} or {'au}.

{'u} is tricky. It would have to be distinguishable from {fu} and {xu}, and both Katakana and Hiragana have none for {'u} and {xu}. フ is actually {fu}, but you have assigned it to {fy}.
 

ワ is pronounced {vy.}, and followed by a vowel, as in vu: ワウ.

ワ is actually {ua}. ワウ is {uau}.

The foreign {v} sound in Japanese is typically transliterated as ヴ, as in ラヴ for "love".

The 'k' kana with º is pronounced {xy.} as in xrula: クºルウラ″ア.

That's an interesting idea. ゜ for a non-labial is never used in formal Japanese. It looks funny to the natives, so it sometimes gets used in manga for comical expressions.

[Jonathan Jones]

On Thu, Sep 2, 2010 at 3:49 PM, tijlan <jbot...@gmail.com> wrote:

On 2 September 2010 20:00, Jonathan Jones <eye...@gmail.com> wrote:

ｼ is pronounced {cy.}, and followed by a vowel or 'y' kana, as in cusku: シウスクウ or シュスクウ.

The latter one should be シュスク (without ウ at the end).

All the "Cu" kana are pronounced without the 'u' bit, as です is pronounced 'des', in the proposal. As such, the ウ has to be there. Otherwise, how would you know that シュスク is pronounced {cusku} and not {cusuku}?

To write {si}, use スイ.

To write {zi}, use ズイ.

スイ is more like {sui}. I suggest スィ. Likewise, ズィ for {zi}.

See above, although the subscript イ is basically the same thing, as far as I'm concerned. (It does look a bit better.)

ジ is pronounced {jy.} and follows the same rules as シ, as in djica: ヅジイシャ.

or ヂシャ. {tci} can unambiguously be チ, and {dji} ヂ.

It has to be that ヂ is  {di}, チ is {ti}, and {tci} is ツシイ, or it would break the rest of the proposal.

For example: Is ヂクル″ {diklu} or {djiklu}?

ヲis pronounced {.y.}

To write {y'u}, use ハウ.

ハウ is most probably {xau} or {'au}.

{'u} is tricky. It would have to be distinguishable from {fu} and {xu}, and both Katakana and Hiragana have none for {'u} and {xu}. フ is actually {fu}, but you have assigned it to {fy}.

 

Because only フ is pronounced with the {fy.} sound, it is {fy.}. So, {fa}, {fi}, {fu},{fe}, {fo} are フア, フイ, フウ, フエ, フオ. Since this means there is no 'hu', I decided on using ハウ to mean that. Doing a subscript ウ would be preferable. I considered ヒウ, ヘウ, and ホウ as well, and ハウ just looks the best to me. I also considered making ヘ be {y'y} instead of {y'e}, which would mean that ハ, ヒ, and ホ aren't used at all.

 

ワ is pronounced {vy.}, and followed by a vowel, as in vu: ワウ.

ワ is actually {ua}. ワウ is {uau}.

The foreign {v} sound in Japanese is typically transliterated as ヴ, as in ラヴ for "love".

I didn't know that. I also have no idea how to make that symbol with a standard U.S. keyboard, so....

The reason I chose ワ for {vy.} is because we have ウア, so ワ isn't needed for it. The ウ" works for me.

The 'k' kana with º is pronounced {xy.} as in xrula: クºルウラ″ア.

That's an interesting idea. ゜ for a non-labial is never used in formal Japanese. It looks funny to the natives, so it sometimes gets used in manga for comical expressions.

{xy.} and {ky.} are closely related, and ガ, etc. are already used for {gV}, so that was my best idea. I know that's not what the maru symbolizes. (It is called maru, yes?)

[Jonathan Jones]

2010/9/2 tijlan <jbot...@gmail.com>

Just out of curiousity, do those kanji in your examples have the meaning of the gismu? Does, for example,  魚 mean fish? I ask because it seems to me that that would make it more difficult to read and write Lojban, because it would be much harder to distinguish from nihongo.

[Jonathan Jones]

On Thu, Sep 2, 2010 at 6:12 PM, Jonathan Jones <eye...@gmail.com> wrote:

On Thu, Sep 2, 2010 at 3:49 PM, tijlan <jbot...@gmail.com> wrote:

On 2 September 2010 20:00, Jonathan Jones <eye...@gmail.com> wrote:

シ is pronounced {cy.}, and followed by a vowel or 'y' kana, as in cusku: シウスクウ or シュスクウ.

The latter one should be シュスク (without ウ at the end).

All the "Cu" kana are pronounced without the 'u' bit, as です is pronounced 'des', in the proposal. As such, the ウ has to be there. Otherwise, how would you know that シュスク is pronounced {cusku} and not {cusuku}?

Or {cusk}, for that matter.

[tijlan]

2010/9/3 Jonathan Jones <eye...@gmail.com>

2010/9/2 tijlan <jbot...@gmail.com>

Some Japanese jbopre have attempted the Katakana transliteration for gismu and cmavo on the jbo-pon dictionary (http://www.editgrid.com/explore/user/tijlan/%E3%83%AD%E3%82%B8%E3%83%90%E3%83%B3%E8%BE%9E%E6%9B%B8) which i host on my EditGrid account.

I myself am a Japanese speaker and have a good command of its orthography. Katakana, like Hiragana, is a syllabary, meaning that it's not cut out for representing single consonants. Every brivla has a consonant cluster, and no Katakana except { ン } [n] can adequately represent that. Cmavo, on the other hand, have no such cluster and can at least by principle be accommodated by a syllbary. More suited for brivla -- content words -- is Kanji, the logograms. For example:

mi carna
ミ　回

ta tricu cmana
タ　木　山

Logograms, when you know them, are a great potential for speed reading.

Katakana, however, has other difficulties in substituting for ASCII. It doesn't distinguish /r/ and /l/, for instance. As you've shown above, { ri } and { li } both become { リ }. { 'a } and { xa } become { ハ }. And so on. One workaround may be to use Hiragana for one of each pair. Say, { ri | り } (Hiragana) & { li | リ } (Katakana), and { 'a | は } (Hiragana} & { xa | ハ } (Katakana}. And certain syllables would have to be represented with two letters, such as { cu | シュ }, which is a combination of { シ } and { ュ }, which is the smaller of { ユ }. It can get quite complicated for non-natives.

moklu gi'e pimlu
口　ギへ　羽

finpe vi lo rirxe midju
魚　ヴィ　ろ　川　中

Just out of curiousity, do those kanji in your examples have the meaning of the gismu? Does, for example,  魚 mean fish?

Yes. And its shape derived from the actual appearance of a fish jumping out of water.

I ask because it seems to me that that would make it more difficult to read and write Lojban, because it would be much harder to distinguish from nihongo.

That might be the case, yes.

The good thing about Kanji, though, is that it's potentially much more intuitive than alphabets and syllabary.

回
シアルヌア
carna

The brain incurs less cognitive process with pictorial/logographic Kanji.

[tijlan]

On 3 September 2010 01:12, Jonathan Jones <eye...@gmail.com> wrote:

All the "Cu" kana are pronounced without the 'u' bit, as です is pronounced 'des', in the proposal. As such, the ウ has to be there.

That's what i understood about the first option, シウスクウ. You add ウ to シ and ク, since they are supposed to lack the /u/ bit, yes.

What appeared to me odd was that, in シュスクウ, ュ is set for /u/, while also ウ is to represent /u/. Also, normally, the small ュ isn't used individually but as a component of a semi-ligature like シュ. So, i thought you meant the second alternative to be more of an alphabetic (phoneme-based) rather than syllabic (syllable-based) solution for {cusku}, where the last ウ would be redundant.

Otherwise, how would you know that シュスク is pronounced {cusku} and not {cusuku}?

You wouldn't. And a workaround i can think of is to use the interpunct inherent to the Japanese input methods:

 cusku -- シュス・ク
 cusuku -- シュスク

ジ is pronounced {jy.} and follows the same rules as シ, as in djica: ヅジイシャ.

or ヂシャ. {tci} can unambiguously be チ, and {dji} ヂ.

It has to be that ヂ is  {di}, チ is {ti}, and {tci} is ツシイ, or it would break the rest of the proposal.

You 1) split {ci} into シイ, but not {ti} into チイ, and 2) set ツ for {t}? Why not split {ti} into チイ so as to make the scheme more consistent?

In your current proposal (sometimes phoneme-based, sometimes syllable-based):

c -- シ
t -- ツ
i -- イ

ci -- シイ
ti -- チ (rather than more predictable ツイ)
cti -- シチ (rather than more predictable シツイ)
tci -- ツシイ

In my suggestion (always phoneme-based):

c -- シ
t -- チ
i -- イ

ci -- シイ
ti -- チイ
cti -- シチイ
tci -- チシイ

 

For example: Is ヂクル″ {diklu} or {djiklu}?

It could be even {dkr}.

For {diklu}, maybe ヂイクリウ.
For {djiklu}, maybe ヂジイクリウ.

 

ヲis pronounced {.y.}

To write {y'u}, use ハウ.

ハウ is most probably {xau} or {'au}.

{'u} is tricky. It would have to be distinguishable from {fu} and {xu}, and both Katakana and Hiragana have none for {'u} and {xu}. フ is actually {fu}, but you have assigned it to {fy}.

 

Because only フ is pronounced with the {fy.} sound, it is {fy.}. So, {fa}, {fi}, {fu},{fe}, {fo} are フア, フイ, フウ, フエ, フオ.

Right, that would be reasonable for a phoneme-based Kana scheme.

Since this means there is no 'hu', I decided on using ハウ to mean that. Doing a subscript ウ would be preferable. I considered ヒウ, ヘウ, and ホウ as well, and ハウ just looks the best to me.

Sounds fine to me too.

 

I also considered making ヘ be {y'y} instead of {y'e}, which would mean that ハ, ヒ, and ホ aren't used at all.

ヘ is a single line and looks a bit counter-intuitive for {y'y}. Why not ホ, which has a visual symmetry analogous to {y'y} as a whole?

 

ワ is pronounced {vy.}, and followed by a vowel, as in vu: ワウ.

ワ is actually {ua}. ワウ is {uau}.

The foreign {v} sound in Japanese is typically transliterated as ヴ, as in ラヴ for "love".

I didn't know that. I also have no idea how to make that symbol with a standard U.S. keyboard, so....

In the Microsoft IME "full-width Katakana" mode, you get ヴ by simply typing "vu".

 

The reason I chose ワ for {vy.} is because we have ウア, so ワ isn't needed for it. The ウ" works for me.

So, {ua} is ウア, and {u.a} is ウ．ア?

The 'k' kana with º is pronounced {xy.} as in xrula: クºルウラ″ア.

That's an interesting idea. ゜ for a non-labial is never used in formal Japanese. It looks funny to the natives, so it sometimes gets used in manga for comical expressions.

{xy.} and {ky.} are closely related, and ガ, etc. are already used for {gV}, so that was my best idea. I know that's not what the maru symbolizes. (It is called maru, yes?)

Sounds fine.

(Yes, it's called "maru" colloquially or "handakuten" linguistically.)

[Jonathan Jones]

My idea behind this is that it would frequently save some space to use the other kana, for instance, ブアシルウ is bacru, but バシルウ saves a character. This is why チ is {ti}, not {t}. Maybe a chart is in order? (The bold entries are the important ones.)

ア - {a}
イ - {i}
ウ - {u}
エ - {e}
オ - {o}

カ - {ka}
キ - {ki}
ク - {ky.}, クウ - {ku}, クア - {ka}, etc.
ケ - {ke}
コ - {ko}

ガ - {ga}
ギ - {gi}
グ - {gy.}, グウ - {gu}, グア - {ga}, etc.
ゲ - {ge}
ゴ - {go}

カº - {xa}
キº - {xi}
クº - {xy.}, クºウ - {xu}, クºア - {xa}, etc.
ケº - {xe}
コº - {xo}

サ - {sa}
シ - {cy.}, シア, シャ - {ca}, シイ - {ci}, シウ, シュ  - {cu}, シェ, シエ - {ce}, シオ, ショ - {co}
ス - {sy.}, スウ - {su}, スイ - {si}, スア - {sa}, etc.
セ - {se}
ソ - {so}

ザ - {za}
ジ - {jy.}, ジア, ジャ - {ja}, ジイ - {ji}, ジウ, ジュ - {ju}, ジェ, ジエ - {je}, ジオ, ジョ - {jo}
ズ - {zy.}, ズウ - {zu}, ズイ - {zi}, ズア - {za}, etc.
ゼ - {ze}
ゾ - {zo}

タ - {ta}
チ - {ti}
ツ - {ty.}, ツウ - {tu}, ツア - {ta}, etc.
テ - {te}
ト - {to}

ダ - {da}
ヂ - {di}
ヅ - {dy.}, ヅウ - {du}, ヅア - {da}, etc.
デ - {de}
ド - {do}

ナ - {na}
ニ - {ni}
ヌ - {ny.}, ヌウ - {nu}, ヌア - {na}, etc.
ネ - {ne}
ノ - {no}

ハ - {y'y.}
ヒ - {y'y.}
フ - {fy.}, フア - {fa}, フイ - {fi}, フウ - {fu}, フエ - {fe}, フオ - {fo}
ヘ - {y'y.}
ホ - {y'y.}

(I believe doing the {hV} series this way causes the least problems. Basically, any {hV} except フ, which is {fy.}, acts as { ' } does in the standard orthography, so ミハア, ミヒア, ミヘア, and ミホア, and variants such as ムイハア, are all {mi'a}. (It also might be better to just keep { ' }, as in ミ’ア. (Or possibly use ー, as in ミーア, which unlike { ' }, works in the vertical as well.)))

バ - {ba}
ビ - {bi}
ブ - {by.}, ブウ - {bu}, ブア - {ba}, etc.
ベ - {be}
ボ - {bo}

パ - {pa}
ピ - {pi}
プ - {py.}, プウ - {pu}, プア - {pa}, etc.
ペ - {pe}
ポ - {po}

マ - {ma}
ミ - {mi}
ム - {my.}, ムウ - {mu}, ムア - {ma}, etc.
メ - {me}
モ - {mo}

ラ - {ra}
リ - {ri}
ル - {ry.}, ルウ - {ru}, ルア - {ra}, etc.
レ - {re}
ロ - {ro}

ラ″ - {la}
リ″ - {li}
ル″ - {ly.}, ル″ウ - {lu}, ル″ア - {la}, etc.
レ″ - {le}
ロ″ - {lo}

ヴ - {vy.}, ヴア - {va}, etc.

ヲ - {y}

Basically, カ is (for an example), in this proposal, merely a way of saving a character to write クア{ky.}{a}.

On Fri, Sep 3, 2010 at 5:25 AM, tijlan <jbot...@gmail.com> wrote:

On 3 September 2010 01:12, Jonathan Jones <eye...@gmail.com> wrote:

All the "Cu" kana are pronounced without the 'u' bit, as です is pronounced 'des', in the proposal. As such, the ウ has to be there.

That's what i understood about the first option, シウスクウ. You add ウ to シ and ク, since they are supposed to lack the /u/ bit, yes.

What appeared to me odd was that, in シュスクウ, ュ is set for /u/, while also ウ is to represent /u/. Also, normally, the small ュ isn't used individually but as a component of a semi-ligature like シュ. So, i thought you meant the second alternative to be more of an alphabetic (phoneme-based) rather than syllabic (syllable-based) solution for {cusku}, where the last ウ would be redundant.

Otherwise, how would you know that シュスク is pronounced {cusku} and not {cusuku}?

You wouldn't. And a workaround i can think of is to use the interpunct inherent to the Japanese input methods:

 cusku -- シュス・ク
 cusuku -- シュスク

ジ is pronounced {jy.} and follows the same rules as シ, as in djica: ヅジイシャ.

or ヂシャ. {tci} can unambiguously be チ, and {dji} ヂ.

It has to be that ヂ is  {di}, チ is {ti}, and {tci} is ツシイ, or it would break the rest of the proposal.

You 1) split {ci} into シイ, but not {ti} into チイ, and 2) set ツ for {t}? Why not split {ti} into チイ so as to make the scheme more consistent?

In your current proposal (sometimes phoneme-based, sometimes syllable-based):

c -- シ
t -- ツ
i -- イ

ci -- シイ
ti -- チ (rather than more predictable ツイ)

({ti} can be either, チ just saves a character in comparison to ツイ.)

Yes, and {u'a} is ウハア, ウ’ア, or ウーア.

The 'k' kana with º is pronounced {xy.} as in xrula: クºルウラ″ア.

That's an interesting idea. ゜ for a non-labial is never used in formal Japanese. It looks funny to the natives, so it sometimes gets used in manga for comical expressions.

{xy.} and {ky.} are closely related, and ガ, etc. are already used for {gV}, so that was my best idea. I know that's not what the maru symbolizes. (It is called maru, yes?)

Sounds fine.

(Yes, it's called "maru" colloquially or "handakuten" linguistically.)

--

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.