Monday 19 July 2021

An initial look at vowel patterns

The first word in my glossary is ab - father, parent, ancestor, etc. It occurs in its base form, root only, 31 times. 30 of those times is carries a full qamats - the sound of 'a'. Once it lowers the full tone of the vowel to patah, a slight decrease in the sound of 'a'. This is the only time that the word 'ab' (without any affixes) occurs in the construct.

So there's a property of aleph, for ab it carries the vowel 'a'.

What other vowels does aleph carry and how often? I have taken these correspondences for vowels. In the following the colon or /-/ indicates a reduction in the vowel sound.

  • schwa, ':',
  • hatef-segol, ':e', 
  • hatef-patah, ':a-', 
  • hatef-qamats, ':a',
  • hiriq, 'i',
  • tsere, 'ei',
  • segol, 'e'.
  • patah, 'a-',
  • qamats, 'a',
  • holam, 'o',
  • qubuts, 'u'.
With 11 variances on the vowel sounds, there is no possibility for a simple presentation. To further investigate the vowels that aleph carries, I ran the algorithm for roots beginning with abd. There are 26 variations among the 62 instances of abd that occur without a prefix. If I reduce the analysis to the base vowel carried by the aleph, it is /a/ 45 times, /i/ 5 times, and o 12 times. So for this guttural, roughly 75% of the time it carries a variation on /a/ when there is no prefix.

This barely scratches the surface. If I allow words with abd as the first three letters of the root and  include wordforms with prefixed letters, the pattern _:_a_ (prefix with schwa and aleph carrying an a) occurs 65 times. And 46 times, there is no /a/ in the vowels at all. 

This is all neither helpful nor presentable! 

What if I test one single prefix at a time: b h v c l m?
b as prefix occurs 4 times, twice on its own and twice with a v becoming u in SimHebrew
_:_a_:_a_ 
_a_a_a_o_ 
u_:_a_:_a_ 
u_a_a_o_ 

See if you can find them in the concordance (abd).

h has too many lines so I eliminated all the ones with suffixes:
_:_a_a_i_ (3 times _a_ in the third position)
_:_e_e_i_ (1)
_a_a_i_ (twice _a_ in position 1)
u_:_a_a_i_ (3 times _a_ in position 4)

v - also complicated. Eliminating the suffixes, I get this:
_:_a_a_ (11)
_:_a_ei_ 
_:_i_a_ (2)
_:_o_ei_ (1)
_:_u_a_ (2)

c never occurs alone as a prefix for this set of roots.
l:
_:_a_:_a_ (7)
_:_a_:_ei_i 
_:_a_ei_ 
_:_o_ei_ (1)
_a_a_a_o_ (1)
u_:_a_:_a_ (4)
u_:_a_ei_ 

m: only two instances
_:_a_:_i_ 
_a_a_i_ 

I think the optimum thing to do is to include some form of the vowels in the concordance [See next post] - they are there in the square text of course, but not so obvious to the 'Roman' reader. Before I do that I will test with a verb form to see what I uncover. There are so many pronoun suffixes where the vowel patterns should emerge clearly.


No comments:

Post a Comment