[YG Conlang Archives] > [saweli group] > messages [Date Index] [Thread Index] >


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

Re: [saweli] Saweli Dictionary



Latejami [L] uses an absolute argument-changing system. 
Saweli-Saxita [S-S] uses a relative argument-changing system. 

In the following table, the L argument-changing suffixes on the top row.
The left column is all of the possible arrangements of the three component elements, viz.: agent-patient [A-P], presence or absence of a focus, and static vs. dynamic. 

In L, to change a word of any argument [arg] structure to e.g., AP-s, just add the suffix "-is". Very simple. 
To change a stem's arg structure in S-S, you have to know the arg structure of the stem first. 
The reason I changed from the absolute system to the relative one is that since L stems have inherent arg structures, you have to know what that is to know when to leave out the arg-structure suffix. The relative system requires knowing the stem's arg structure, and figuring out which suffix to add to change the three elements of the is simply applying the small table of elements shown below. 

The absolute system only gives the end-point, so to speak, with no regard to the starting point. The relative system relies on knowing the starting point and gives the changes to the structure necessary to get to the desired end point, which is always recoverable, unlike the L start point, which is lost (irrecoverable) as soon as an arg-structure suffix is added.  

In the following table, the entries "(vi)" are for unchanged arg structures, not meant to be used in practice since they're totally redundant. 

To change the arg structure from P-s to P/F-s, e. g.,  in L, add -unz. To accomplish the same in S-S, you add "wi". The "w" means "add focus", and the "i" means "no change in A-P". The difference becomes even more apparent when you realize that L "unz" makes any stem in L P/F-s, whereas in S-S, the suffix "wi" is used for all the changes from unfocused to focused, with no other changes. 
FYI: each L suffix is decomposable into elements as well. u/i/a are for A-P, n/m is for focus, and z/s vs. b/p is for dynamicity. 
from>to us unz is inz as anz up umb ip imb ap amb
P-s (vi) wi vo wo vu wu zi zwi zo zwo zu zwu
P/F-s yi (vi) yo vo yu vu zyi zi zyo zo zyu zu
AP-s ve we (vi) wi vo wo ze zwe zi zwi zo zwo
AP/F-s ye ve yi (vi) yo vo zye ze zyi zi zyo zo
A/P-s va wa ve we (vi) wi za zwa ze zwe zi zwi
A/P/F-s ya va ye ve yi (vi) zya za zye ze zyi zi
P-d xi xwi xo xwo xu xwu (vi) wi vo wo vu wu
P/F-d xyi xi xyo xo xyu xu yi (vi) yo vo yu vu
AP-d xe xwe xi xwi xo xwo ve we (vi) wi vo wo
AP/F-d xye xe xyi xi xyo xo ye ve yi (vi) yo vo
A/P-d xa xwa xe xwe xi xwi va wa ve we (vi) wi
A/P/F-d xya xa xye xe xyi xi ya va ye ve yi (vi)
In the following chart, the changes to the original argument are shown for each piece of information in the structure (dynamicity, focus, agent-patient).  

s/d                 +/- F                         P|AP|A/P

x=>s               w=>+F                       a = 2 lower [A/P > P]

z=>d               y=>-F                        e = 1 lower [AP > P, A/P > AP]

0|v=no chng    0=no chng.                i = no change

                                                      o = 1 higher [P > AP, AP > A/P]

"v" is used when only the                  u = 2 higher [P > A/P]

vowel changes, to provide a consonant.


I originally simply copied the L way into S-S, but kept hitting the lost info problem. The relative system solves that. 


This is a lot of info and possibly not a very clear explanation, so ask me if you have more questions. 


stevo



On Wed, Nov 22, 2017 at 3:07 AM, geoff.hacker@hidden.email [saweli] <saweli@yahoogroups.com> wrote:
 

Cool. So, what does it mean to have a relative argument-changing system, as opposed to an absolute one?