cmevla–brivla merger: Difference between revisions

From Lojban Wiki
Jump to navigation Jump to search
mNo edit summary
m (i'm unsure as to whether this was the intended page for this link, but if it isn't, better this than nothing.)
 
(3 intermediate revisions by one other user not shown)
Line 1: Line 1:
The '''Cmevla–Brivla Merger''' {{editorial|Lojban name?}} (often abbreviated as '''CBM''') is a [[Modern Lojban|common addition to the grammar]] that treats cmevla (name words, words ending in a consonant) as a type of brivla (predicate words).
The '''Cmevla–Brivla Merger''' {{editorial|Lojban name?}} (often abbreviated as '''CBM''') is a [[Modern Lojban concepts|common addition to the grammar]] that treats cmevla (name words, words ending in a consonant) as a type of brivla (predicate words).


== History ==
== History ==
Line 18: Line 18:
* {{j|.tag.}}: {{x|1}} <code>@</code>-s {{x|2}} in message {{x|3}}.
* {{j|.tag.}}: {{x|1}} <code>@</code>-s {{x|2}} in message {{x|3}}.


(These definitions are informal; therefore, one may expect others to be confused with their usage of the oblique places unless it’s rather obvious – e.g., ''I chessed her and she lost'' {{j|mi ko’a .tces .i je ko’a te jinga}}.)
(These definitions are informal; therefore, one may expect others to be confused with their usage of the oblique places unless it’s rather obvious – e.g., ''I chessed her and she lost'' {{j|mi ko'a .tces .i je ko'a te jinga}}.)


Therefore, CBM cmevla, alongside the underratedly versatile {{dic|me'oi}}, can find great use when code-switching ({{j|lo nunctu zo'u '''.borin''' .oi sai}}) or learning ({{j|mi klama '''lo .skul.'''}}).
Therefore, CBM cmevla, alongside the underratedly versatile {{dic|me'oi}}, can find great use when code-switching ({{j|lo nunctu zo'u '''.borin''' .oi sai}}) or learning ({{j|mi klama '''lo .skul.'''}}).
Line 31: Line 31:


<blockquote>
<blockquote>
   {{j|.i mi ba’o penmi ci lo .alis.}} <br />
   {{j|.i mi ba'o penmi ci lo .alis.}} <br />
   ''I’ve met three Alices so far.''
   ''I’ve met three Alices so far.''
</blockquote>
</blockquote>
Line 95: Line 95:


Notice how in zantufa (which admits CBM), cmevla (represented by the production rule <code>CMEVLA_clause</code>) are capable of serving as <abbr title="Units in a tanru. «condi» and «me lo bloti» are one tanru unit each; «condi me lo bloti» is two.">tanru units</abbr> to the same extent that brivla (<code>BRIVLA_clause</code>) are.
Notice how in zantufa (which admits CBM), cmevla (represented by the production rule <code>CMEVLA_clause</code>) are capable of serving as <abbr title="Units in a tanru. «condi» and «me lo bloti» are one tanru unit each; «condi me lo bloti» is two.">tanru units</abbr> to the same extent that brivla (<code>BRIVLA_clause</code>) are.
Now let’s examine selma’o {{j|LE}} and {{j|LA}}. (In both cases, <code>sumti_tail</code> depends on <code>selbri</code>, which itself is one or more <code>tanru_unit</code>s, defined above.)
'''camxes'''
sumti_6 <- [...] / LA_clause free* relative_clauses? CMEVLA_clause+ free* / (LA_clause / LE_clause) free* sumti_tail KU_elidible free* / [...]
'''zantufa'''
sumti_5 <- [...] / LE_clause sumti_tail KU_elidible / [...]
camxes defines cmavo of {{j|LA}} as cmavo that can be followed both by a cmevla and a brivla, and {{j|LE}} as those that can only be followed by the latter. Since zantufa is a CBM grammar, it doesn’t make the distinction; thus, it doesn’t need to define the selma’o {{j|LA}} (as well as the rule <code>LA_clause</code>) at all.

Latest revision as of 17:34, 8 November 2023

The Cmevla–Brivla Merger [Lojban name?] (often abbreviated as CBM) is a common addition to the grammar that treats cmevla (name words, words ending in a consonant) as a type of brivla (predicate words).

History

[TODO]

Usage

Under CBM, each and every cmevla is treated as a predicate. But what kind of predicate? The most valid (though generic) answer to this question is: whatever predicate is named by the cmevla in question. For example, names of people, places, countries, etc., may simply be ‘predicatized’ like so:

  • .alis.: x1 is somebody/something named Alice; x1 is [an] Alice.
  • .djonguuen.: x1 is somebody/something named .djonguuen.; x1 is [a] Chinese language.
  • .amerikys.: x1 is somebody/something named America; x1 is [an] America.

But in cases where it matters – like when we’re naming an action or a more complex relation – we may give the cmevla predicate any structure that we like:

  • .tces.: x1 plays chess with x2.
  • .tag.: x1 @-s x2 in message x3.

(These definitions are informal; therefore, one may expect others to be confused with their usage of the oblique places unless it’s rather obvious – e.g., I chessed her and she lost mi ko'a .tces .i je ko'a te jinga.)

Therefore, CBM cmevla, alongside the underratedly versatile me'oi, can find great use when code-switching (lo nunctu zo'u .borin .oi sai) or learning (mi klama lo .skul.).

Hold on now, is it la .alis. or lo .alis.?

Both forms are correct, mean different things, and belong to different contexts of usage.

la .alis. means one who is called Alice, just like la rokci is understood to mean one who is called Rock.

lo .alis. means an Alice, as in:

.i mi ba'o penmi ci lo .alis.
I’ve met three Alices so far.

[TODO – expand?]

Tanru with CBM

Observe:

.i la citno .alis. cu zutse ne'a lo rirxe korbi
.i la .alis. co citno cu zutse ne'a lo rirxe korbi
Young Alice sat beside the river bank.

.i .ia nai dai mi na'e ju'o se slabu lo .boloniiéz. te jukpa
Unbelievable, I know, but I really have no idea how to cook spaghetti bolognese!

Caveat: la syntax

Without CBM, a LA phrase terminated as soon as there were no more cmevla – for example:

.i la .alis. citno
Alice is young. (CLL Lojban)

With CBM, though, the grammatical reading of the sentence diverges:

.i la .alis. citno
Alice-like Youngster (under CBM)

This is because .alis. and citno now form an accidental tanru, meaning (to be) Alice type-of young. Since la behaves exactly the same as lo and le and all the other LE with the introduction of CBM, one must take care to splice in a ku or cu to prevent the tanru from forming:

.i la .alis. cu citno
Alice is young. (with and without CBM)

※ CLL LA does not take a ku when followed by a cmevla (why would it need to if it knows when to terminate?), so this sentence

.i la .alis. ku citno
Alice is young.

will not parse without CBM.

Formal description

Compare the following snippets from camxes’ and zantufa’s grammar files:

camxes

tanru_unit_2 <- BRIVLA_clause free* / [...]

zantufa

tanru_unit_1 <- (CMEVLA_clause / BRIVLA_clause / [...]

Notice how in zantufa (which admits CBM), cmevla (represented by the production rule CMEVLA_clause) are capable of serving as tanru units to the same extent that brivla (BRIVLA_clause) are.

Now let’s examine selma’o LE and LA. (In both cases, sumti_tail depends on selbri, which itself is one or more tanru_units, defined above.)

camxes

sumti_6 <- [...] / LA_clause free* relative_clauses? CMEVLA_clause+ free* / (LA_clause / LE_clause) free* sumti_tail KU_elidible free* / [...]

zantufa

sumti_5 <- [...] / LE_clause sumti_tail KU_elidible / [...]

camxes defines cmavo of LA as cmavo that can be followed both by a cmevla and a brivla, and LE as those that can only be followed by the latter. Since zantufa is a CBM grammar, it doesn’t make the distinction; thus, it doesn’t need to define the selma’o LA (as well as the rule LA_clause) at all.