Re: Exciting New Discovery!

[Jacques Guy <jguy@xxxxxxxxxxxxxxxx>]

Bruce Grant wrote:
> 
> Even if the _ch_ character represents "and", it might not represent the letters
> "e t" if the underlying language isn't Latin.

If EVA ch is "et", then we have a language where t can only be preceded
by either of two letters: "e" and whatever EVA s is. True, seldom, very
seldom (once in five blue moons) you find the sequence ih. It still 
remains very, very unlike Latin, or any European language I can 
think of.

Now if ch means "and", but is not read "et", the pattern remains
the same, and fits no language I can think of. 

Look now. Consider these three letters of the Roman alphabet:

h, n, and m

h is l plus what I'll call  X (the remainder)
n is dotless i plus X
n is dotless i plus X plus X

This "letter" X can only be preceded by itself, or l,
or dotless i. A very restrictive distribution, isn't it.
(And it never occurs isolated either).
Its cause? The misanalysis or our Roman alphabet: mistaking
strokes for letters, elements of letters for whole letters.
So with ch and sh. In my view, they are single letters,
or perhaps even elements of single letters. They are
emphatically not sequences of two letters.