reduced logical form: Difference between revisions

From Lojban
Jump to navigation Jump to search
mNo edit summary
 
mNo edit summary
Line 1: Line 1:


Names of languages in Lojban have three forms:
sentence = selbri-2 [[jbocre: FA|FA]] variable] ... /VAU/


*a gismu, type=4 fu'ivla, or sometimes lujvo with the place structure ''x1 is fooish in aspect x2'', which can also be used for culture or nationality, depending on the language
| gek sentence gik sentence
*a lujvo in ''-bau'' or type-3 fu'ivla with the place structure ''x1 is the fooish language used by x2 to express x3''


*a cmene.
| NA KU ZOhU sentence


==== Indo-European ====
| tag /KU/ ZOhU sentence


*Germanic
| [[jbocre: tag|tag]] [[jbocre: quantifier|quantifier]] variable [[jbocre: relative-clauses|relative-clauses]] ZOhU sentence
**English


***glico
variable = KOhA [[jbocre: XI number|XI number]]  (da, de, di; ko'a, ko'e, ..., fo'u)
***glibau


***gliban
selbri-2, gek, gik, tag, quantifier, relative-clauses, number as in the EBNF grammar
***bangrglici


**Scots
The above grammar is a sub-grammar of the Lojban grammar. By that I mean that every sentence it generates is a valid sentence of the Lojban grammar. I contend that every Lojban sentence has a corresponding reduced logical form generated from this grammar. The idea is simple: to get the reduced logical form eliminate indicators and free modifiers, move quantifiers and negations to the prenex, expand all logical connectives to forethought sentence connectives and finally move all the arguments behind the selbri. The devil is, of course, in the details. I intend to work out an algorithm for producing the reduced logical form for any Lojban sentence. (In fact I think there are a couple of tricky places where it may not be doable, but at least this will show what they are.) The purpose of this exercise is to clarify the argument for the scope of NA. In the reduced logical form the relative scopes of the different operators are fairly obvious. mi'e [[User:xorxes|xorxes]]
**Frisian


***bangrfrizi
=== Algorithm  ===
**Dutch


***bangrnederlanda
to reduce any Lojban text to its logical form as defined in the above grammar (Work in progress)
**Afrikaans


**German
[[jbocre: Reduced logical form: Step 1| Step 1]]: Reduce ''text'' to ''paragraphs''
***dotco


***dotybau
[[jbocre: Reduced logical form: Step 2| Step 2]]: Reduce ''fragment'' to ''statement''
***dotyban


**Yiddish
[[jbocre: Reduced logical form: Step 3| Step 3]]: Reduce ''statement'' to ''subsentence''
***brodo'o ''(this might sound depreciatory: "Judendeutsch")''


***iidic
[[jbocre: Reduced logical form: Step 4| Step 4]]: Reduce ''paragraphs'' to ''subsentence''
**Danish


***bangrdanska
[[jbocre: Reduced logical form: Step 5| Step 5]]: Reduce ''bridi-tail'' to ''bridi-tail-3''
**Swedish


**Norwegian
[[jbocre: Reduced logical form: Step 6| Step 6]]: Reduce afterthought sumti and termset connectives to forethought, and quantified terms to KOhA1-form.
**Faroese


**Icelandic
[[jbocre: Reduced logical form: Step 7| Step 7]]: Reduce pre-selbri terms to KOhA
*Celtic


**Gaelic (''is this Irish or Scottish Gaelic?'')
[[jbocre: Reduced logical form: Step 8| Step 8]]: Reduce ''bridi-tail'' to ''selbri-6'' and KOhA. Obtain final form.
***skoto (''it's funny having this gismu but none for Irish!'')


***ga'elo (?) ''that breaks into ga'e lo'' '''No. ga'elo is ga'Elo, while ga'e lo is gA'elo or gA'elO, so it is ok.''' ''It is not OK. A brivla must have two adjacent consonants, which this doesn't. And ga'e lo can be stressed ga'E lo.''
This reduction does not handle:
***gailge


***goidle (''Goidelic'')
# Indicators and free modifiers
***bangrxeire


***gailban
# VUhO relative-clauses
**Cymru


***kemru, kemro
# tags when they double up with other connectives
***kemrybau
 
***kemruban, kemryban
**** Is it worth noting that Welsh "Cymru" is pronounced pretty much like Lojban '''kymri''' and not '''kemri''' or anything ending in '''u'''? ''You can't have '''y''' in gismu, and all cultural gismu end with '''o''', so it should be '''kemro'''.''  OK, then it should at least be ''kimro'', since the Welsh '''y''' also has a "clear" sound like Lojban '''i'''.
 
***** Welsh has three separate phonemes involved here: y, i, and u. I is an unrounded lojban i. U is a rounded i (except in some southern dialects, where it has merged with i). Y varies; it is a lojban y in some monosyllabic words and in all non-final syllables, but elsewhere it is identical to u. Lojban only having five vowels for us to use where welsh has 7 (a,e,i,o,u,w,y), we are forced to merge some. We must merge u and i, because i is the lojban vowel that can be pronounced as a welsh u, while our u is taken (by welsh w). We then must merge y with something. If we merge it with i/u, then we get three mergers (i/u, i/y, and u/y). If we merge it with something else (call it *), we only get two mergers (i/u, y/*). This is clearly preferable. As for what * should be, I would vote for a, as it is the only low vowel we can choose. If the y were in a final syllable, where it is not distinct from u even in welsh, I would vote for using i. But, as a neogism, I would vote for kamro (rather than kamri because all other cultural gismu end with o). However, as a name, kymraig or kymra,eg is probably best. By the way, I am not just talking off the top of my head - what I say here (except for the roundedness of u, which is mentioned elsewhere but is not true of the southern welsh dialects) is backed up by the Welsh Studies Institute in North America, one of the leading authorities on this issue. see [http://www.madog.org] for more detail about the funky welsh language. - mi'e kreig.daniyl.
******You're absolutely right. Although I proposed _kemru/kemro_ myself (because of its gismu-structure), I don't like it any longer for its derivation from _Cymru_ (Cymru [[jbocre: f.|f.]] - (prop. n.) Wales!!). The language is _cymraeg_ - pronounced pretty like the lojban form _kymraig_ below. (_cymreig_ adj. stands for pertaining to welshness) -- mi'e .aulun.
 
***kymraig
**Breton
 
***bre'one
*Italic
 
**Latin
***latmo
 
***la'orbau
***la'orban
 
**Portuguese
***porto
 
***potybau
***potyban
 
**Spanish
***spano
 
***sanbau
***sanban
 
**Catalan
***katlana
 
***bangrkatalana
***katalan ''invalid cmene''
 
***bangrvalenciana
**French
 
***fraso
***fasybau
 
***fasyban
**Italian
 
***talno
***talnybau
 
***talnyban
**Rumanian
 
*Greek
***xelso
 
***xesybau
***xesyban
 
*Slavic
**Russian
 
***rusko
***rukybau
 
***rukyban
**Ukrainian
 
***vukro
**Old Church Slavonic
 
***dzeru'o
***dzeru'obau
 
***dzeruk
*Indic
 
**Hindi
***xindo
 
***xinbau
***xinban
 
**Urdu
***xurdo
 
***xurbau
***xurban
 
**Sanskrit
***srito
 
***sritybau
***srityban
 
**Romani
***tsingaro ''Why on earth take it from Italian ("zingaro") and not from Romanian ("tigan") or Hungarian ("cig�ny") which is more common and typical? -> {tsigano}'' ''ok with me''
 
***bangnromani
***romanes
 
==== Finno-Ugric ====
 
*Hungarian
**magjaro (''madjaro'' breaks up) - ''accepted!''
 
**bangrmagiaro
**madjar
 
==== Basque ====
 
*Basque
**skalduna
 
==== Afro-Asiatic ====
 
*Arabic
**xrabo
 
**rabybau
**rabyban
 
*Hebrew
**xebro
 
**brobau
**broban
 
==== Austronesian ====
 
*Malayo-Polynesian
**Malay and Indonesian
 
***baxso
***baxsybau
 
***baxsyban
**Polynesian
 
***Niuean
****bangrniu,e
 
*Japanese
**ponjo
 
**po'orbau (ponbau)
**po'orban (ponban)
 
==== Sino-Tibetan ====
 
*Chinese
**jungo
 
**jugbau
**jugban
 
==== South Caucasian ====
 
*Georgian
**kartuli
 
**bangrkartuli
 
==== Artificial ====
 
*Loglan offshoots
**TLI Loglan
 
***loglo
***dzelojbo
 
***purjbo
***loglan
 
**Lojban
***lojbo
 
***jbobau
***lojban
 
**[[gua\spi]]
***gu'aispi ''this is not a valid word as there are too many vowels before the first consonant pair'' '''Ok then, guaspi - which isn't a lujvo as ''ua'' cannot appear in rafsi, or at least doesn't. Or we could just give up on those crappy [[jbocre: type 4 fu'ivla]]...'''
 
***bangrguaspi
***guasp
 
**Ceqli
***tcenlis
 
*[[jbocre: Esperanto|Esperanto]]
**spero
 
**bangrsperanto, bangrxesperanto
**esperanton
 
*Laadan
**bangrladanu, banrla,adanu
 
**ladan ''not valid'', ledan, LE,adan
*[[jbocre: Klingon|Klingon]]
 
**bangrtlenanu
**bangrtlingana
 
**tlinan
*[[jbocre: Ro|Ro]]
 
**bangnro
*[[jbocre: Solresol|Solresol]]
 
**solresol
**bangrsolresolo
 
**solresolo
**zgibau
 
**zgiban
*[[jbocre: Sona|Sona]]
 
**son
**bangrsona
 
**djunu'i
 
==== Signal ====
 
*ASL
**merxanbau
 
**mersniban
 
For programming languages, see [[jbocre: Names of Computer Languages|Names of Computer Languages]].

Revision as of 17:09, 4 November 2013

sentence = selbri-2 FA variable] ... /VAU/

| gek sentence gik sentence

| NA KU ZOhU sentence

| tag /KU/ ZOhU sentence

| tag quantifier variable relative-clauses ZOhU sentence

variable = KOhA XI number (da, de, di; ko'a, ko'e, ..., fo'u)

selbri-2, gek, gik, tag, quantifier, relative-clauses, number as in the EBNF grammar

The above grammar is a sub-grammar of the Lojban grammar. By that I mean that every sentence it generates is a valid sentence of the Lojban grammar. I contend that every Lojban sentence has a corresponding reduced logical form generated from this grammar. The idea is simple: to get the reduced logical form eliminate indicators and free modifiers, move quantifiers and negations to the prenex, expand all logical connectives to forethought sentence connectives and finally move all the arguments behind the selbri. The devil is, of course, in the details. I intend to work out an algorithm for producing the reduced logical form for any Lojban sentence. (In fact I think there are a couple of tricky places where it may not be doable, but at least this will show what they are.) The purpose of this exercise is to clarify the argument for the scope of NA. In the reduced logical form the relative scopes of the different operators are fairly obvious. mi'e xorxes

Algorithm

to reduce any Lojban text to its logical form as defined in the above grammar (Work in progress)

Step 1: Reduce text to paragraphs

Step 2: Reduce fragment to statement

Step 3: Reduce statement to subsentence

Step 4: Reduce paragraphs to subsentence

Step 5: Reduce bridi-tail to bridi-tail-3

Step 6: Reduce afterthought sumti and termset connectives to forethought, and quantified terms to KOhA1-form.

Step 7: Reduce pre-selbri terms to KOhA

Step 8: Reduce bridi-tail to selbri-6 and KOhA. Obtain final form.

This reduction does not handle:

  1. Indicators and free modifiers
  1. VUhO relative-clauses
  1. tags when they double up with other connectives
Retrieved from "https://mw.lojban.org/index.php?title=reduced_logical_form&oldid=75889"