Re: [lojban] fu'ivla

[Pierre Abbat <phma@phma.optus.nu>]

On Saturday 10 September 2011 17:26:47 Bob LeChevalier, President and 
Founder - LLG wrote:
> If you are a human being, given that human beings read focusing on the
> starting and ending of words, being able to classify the word as a
> brivla and not a compound cmavo written as one word (like "lenu")
> without having to exercise a complex algorithm, is a virtue.

I think that your supposition is incorrect. In Indo-European languages it is 
generally true that grammatical information is found at the end of words, and 
in Bantu languages it is found at the beginning. In Afro-Asiatic languages, 
though, roots consist of consonants and much of the grammatical information 
is encoded in the vowels between the consonants and in the doubling of 
consonants. One of these languages, namely Arabic, is a source language. And 
many East Asian languages have little if any inflection and consist mostly of 
monosyllabic morphemes. Among them is another source language, Chinese.

Anyone who grows up with one of these morphologies internalizes it. Likewise a 
verclijbo internalizes the rule that the second consonant in a brivla either 
is next to another consonant or is followed by "y" and a consonant. He also 
internalizes the morphology of lujvo and the valid consonant clusters.

If the current entries are any indication, the number of fu'ivla beginning 
with at least three vowels, possibly with apostrophes, and preceded by zero 
or one consonant, is going to be small even if the rule is relaxed. The 
current entries, plus words in my head, beginning with three vowels are:
aierne (and variant ai'erne, to satisfy camxes)
uaizdo, variant of aizdo
io'imbe.
There are several experimental cmavo of the form CV'VV or CVV'V, but none have 
been entered with at least three vowels and no consonant. ("a'oi" isn't 
entered; a nautical vocative, as far as I know, exists only in Germanic 
languages and languages which have borrowed from them.) Any word whose first 
vowel is in a group of at least three, and which ends in a vowel, is either 
an experimental cmavo or a fu'ivla, so the reader will have to think to 
figure out the meaning, which will take much longer than figuring out if it's 
a cmavo or a fu'ivla.

Lastly, a complex algorithm is better than no algorithm at all. In all the 
natlangs I know, there are words that belong to at least two open grammatical 
classes, and I can exhibit such words in three of the six source languages 
(entrance, intimidad, попугаю; all are both nouns and verbs). In Lojban one 
can decide in polynomial time whether something that looks like a brivla is a 
valid brivla, a brivla preceded by a cmavo, or a slinku'i.

Pierre
-- 
Don't buy a French car in Holland. It may be a citroen.

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@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.