Morphological Unrevolution

[trans]

This proposal presents one possible way to simplify the morphology. It is a simplification of the GMR (Great Morphological Revolution). The "Great Morphological Revolution" which introduced affixes, the "djifoa"/"rafsi", and the construction of compound words to the language, may have been seen as revolutionary at the time, but in retrospect it seems a rather overwrought solution. It made the morphology much more complex and less consistent, added schwa hyphens, and required the memorization of whole new set of "mini-words". This article gives a good overview of the issues.

What is propose here instead, is a much simpler system where words are strictly combinations of CVC and CCV followed by CV. For example, all three-syllable words are of the form CVCCVCCV, CCVCVCCV, CVCCCVCV or CCVCCVCV. There is no y (schwa) to represent a hyphen. Nor are there any rafsi to memorize. An advantage of this restriction is that words can be parsed without depending on stress and speech would have more regular rhythm.

Compound words would be formed using the the following patterns, given in the order of precedence.

First Word	Second Word	Combination
CCVcv	CCVCV	CCV,CCVCV or CCV,CCVcv
CCVcv	CVCCV	CCV,CVCCV or CCV,CVCcv
CVCcv	CCVCV	CVC,CCVCV or CVC,CCVcv
CVCcv	CVCCV	CVC,CVCCV or CVC,CVCcv
CCVCV	CCVcv	CCVCV,CCV
CVCCV	CCVcv	CVCCV,CCV

(Note ths later are removed b/c of comment below.)

Some examples:

bitma + salfa => bitsalfa 
bitma + stafa => bitstama

If none of these combinations prove legal or if the word happens to already be an official word in the dictionary (having been constructed from other words), then the alternative is to form a long compound by combining the two words in their entirety and inserting an additional constant between them. For example, bitma and salfa can become bitmarsalfa. The constant chosen is essentially arbitrary, but it should either derive from the original borrowing of the root words, be a doubling of the preceding letter (if allowed, see #), or taken from a common table (still under construction).

nb  tc  zd  f  g  dj  k  l  m  n  mp  r  rs  st  v  x  z

Using this table has the advantage of creating a reoccurring constant pattern, which makes it easier for listeners to parse these long words.

The upshot of this proposal is that all root words have an affix composed of their first three letters, so there is nothing more to learn, and composition of compound words is dead simple. Of course, this also means that a dozen or so words can share the same affix, but more often not it is not an issue. Most words do not go together often enough to ever need compounding. There are however a few words that are often used to form compounds, and in these select cases it may be prudent to devise an additional technique.

Note that multi-syllable borrowed words will not necessarily have a semantic deconstruction. For example balkonni might mean "balcony" and not "balanced cone". Most words of course do deconstruct nicely, just not all.

[trans]

There is an issue with this proposal, a couple of the formation rules skew the intended sematics of the compound word. e.g.

bitma + stafa => bitmasta

Could me mistaken for a compund of bitma and masta, instead of the intended bitma and stafa. So these formations must be removed.

[trans]

I think we can generalize the rules to:

S1s1 + S2s2 = S1S2s1 or S1S2s2

Where S is a syllable with two constants (CVC or CCV), and s is syllable with one constant and ending in a vowel (CV).