Le samedi 8 juillet 2017 22:07:17 UTC, Vincent Broman a écrit :
coi guskant
I agree that part of PA3 is reducible.
The "fihu" and "cehi" are not very productive, and several others could be dispensed with, but at a cost in verbosity.
Peano showed how all arithmetic could be reduced to "pa", "sumji", and "pilji", and the later two could be reduced to just "baihei".
Still I'm disposed to see the convenience of most of PA3 for practical purposes in constructing numbers.
The distinction between common functions in VUhU and more technical functions in brivla preserves the old grey area where you debate about how common or how technical each cmavo is, but you still have to support both kinds of functions separately.
I wished to push the limits, to see how much can be done with the least means.
Every part of speech has a complexity cost and has to justify the cost with expressive productivity.
VUhU looked disposable.
Vincent Broman mihe bremenli
Yes, VUhU and PA3 are disposable. I appreciate using bridi math for all expressions of math. VUhU math is still worth preserving especially when it is applied to non-math objects.
Let us consider the following examples parsed by an experimental grammar.
[1. VUhU math in non-math sumti]
la galaksis cu gunma lo gei papa vei repimu suhi jonai vuhu papimu tarci .i
[2. VUhU math for counting non-math objects]
lo tarci cu gei papa vei repimu suhi jonai vuhu papimu mei .i
[3. bridi math in non-math sumti]
la galaksis cu gunma lo tarci noi zilkancu lo nihai dektefpihi lo nihai sumji jonai se sumji li repimu li papimu kei li papa .i
[4. bridi math for counting non-math objects]
lo tarci cu zilkancu lo nihai dektefpihi lo nihai sumji jonai se sumji li repimu li papimu kei li papa
#note 1. The above statements are parsed by zantufa (http://guskant.github.io/gerna_cipra/zantufa-1.9999.html ) only. The official grammar of mekso doesn't permit even those natural expressions.
#note 2. {zilkancu SUMTI} in the examples of bridi math may be replaced by {me SUMTI (mehu) mei}, just like {zilkancu li PA} that may be replaced by {PA mei}. The definition of {PA mei} and its difference from {zilkancu li PA} are mentioned here: https://mw.lojban.org/papri/gadri:_an_unofficial_commentary_from_a_logical_point_of_view#Inner_quantification .
Syllable count:
VUhU _expression_: 47
bridi _expression_: 79
ratio: 0.59
Discussions.
1.
Besides the syllable count, the important thing for the language being practical is the simpleness of structure that will help listeners' understanding of the meaning. The bridi math requires three layered sub-clauses in the example 3, two layered sub-clauses in the example 4, while the VUhU math requires only one {vei} clause both in the examples 1 and 2. The difference comes from the fact that VUhU math {gei papa vei repimu suhi jonai vuhu papimu} is grammatically separated from the non-math parts of the statements, while bridi math is naturally merged in the non-math structure.
Using the bridi math is quite meaningful for expressions of pure math. However, more layers of sub-clauses are used, more difficult for listeners to grasp the structure of the statement. That annoyance is remarkable especially in non-math statements.
2.
{zilkancu SUMTI} in the example 3 and 4 may be replaced by {me SUMTI (mehu) mei}, but the number of layers will be augmented by {me ... mehu}. The VUhU structure of the official grammar has the same annoyance because it does not permit VUhU math in MOI structure without {me li ... (loho mehu)}.
However, as I have already pointed out in the last message, the grammar around VUhU can be much simpler as shown by Zantufa grammar. If you accept that simplification, then the VUhU math can be harmonized with MOI structure as shown in the example 2, which seems to me not so abominable as the official VUhU math.
mihe la guskant