[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lojban] Re: Lojban-English full dictionary effort- request for assistance

Jorge Llambías wrote:

On 1/3/08, Bob LeChevalier <lojbab@lojban.org> wrote:


I suspect that there aren't many yet because there aren't all that many
lujvo with agreed upon places.


For this question, only the x1 place is relevant.
If there is no x2 place defined for xxxyyy then the selkemxxxyyy lujvo has no obvious meaning. 


(I don't actually know what the count is
for lujvo in jbovlaste that a) are not themselves based on SE b) have
place structures which have seen any usage of non-x1 places.  Of course,
I am not even sure how many 2 part and 3 part lujvo there are in
jbovlaste with defined places structures at all.)
At least tens. 

That's about what I expected.


If among tens of selxxxxxx lujvo, all are (selxxx)xxx and
not one of them has an x1 that corresponds to sel(xxxxxx), that shows
a tendency.
Statistically invalid. There are 1300 gismu. If mere tens have been used to make meaningfully multi-placed lujvo, that is a couple of percent. Now we are trying to conclude how many of those lujvo will be used in meaningful compounds, and I see no reason why we would expect more than a couple of percent. A couple of percent of "tens" could easily be less than 1. But even that is too large an estimate, because you have to eliminate those possibilities where selkemxxxyyy is meaningfully different from xxxselyyy, and you have to have the word useful enough that someone will bother making a lujvo from it rather than merely doing "se xxxyyy"
Meanwhile, I just looked in one of my old lujvo files, I think Nick's analyzed file: 
selkemspadji
terkemselgeimau
are both in there with defined place structures. The former is a weaker example because spadji is not in the list, whereas selgeimau is in the list. 

I have no idea if these words ever saw usage or made it into jbovlaste.
But if you insist on more than a statistical example, or one of unknown usefulness:
xismalsi is clearly a Christian church and I'll arbitrarily assume that it is sufficient to have xriso eliminating the need for an x2 from malsi, for a place structure of 

x1 is a Christian church serving area/parish x2
We would probably want to add a place for the denomination, either before or after the x2. For purposes of argument (I'm doing this off the top of my head) I'll let the denomination be x3. 

Thus selkemxismalsi is English "parish"
and terkemxismalsi is English "denomination".
These naturally stem from having xismalsi being a lujvo with a meaningful place structure, and especially with the x3 that is not predictable from jvojva.
Each of these words is used by those who discuss religion in compounds like "parishioners", "parish priest", and non-compounds of hidden semantics like "vicar", "deacon" and other church offices organized around a local church. The Catholics have another set of offices organized around the diocese level and another at the archdiocese level, and probably at least some would rely on whatever the place structure is for the Lojban word for "diocese".
I suggest that a different set of clusters of words might develop around the multi-layered levels of government in a large country, where we have gismu for city and country, but not for district, county, state/oblast/department. The lujvo for the levels of government will be heavily used in describing the functions and officers of that level of government. And if only x1 is defined, those lujvo will be long and inefficient and non-Lojbanic.
I suspect that a lot of the endless discussion of computer terminology in Lojban comes about because earlier words were not defined including place structures that would automatically cover a few computer concepts with one word. 

Similarly with linguistic terminology (I note that
These lujvo will not exist until a) people want to talk about the subject matter, b) a place structure for the simpler term is settled, c) people are sufficiently confident about their lujvo making competence to bother making lujvo 

lojbab



To unsubscribe from this list, send mail to lojban-list-request@lojban.org
with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if
you're really stuck, send mail to secretary@lojban.org for help.