Re: LSC sums for monkey texts

[Jacques Guy <jguy@xxxxxxxxxxxxxxxx>]

I am trying to think of properties of language,
real languages, and construct falsiable theories.

Take those languages where you cannot have two
consonant in a row, and where a consonant must
always be followed by a vowel. Almost all 
Polynesian languages, many Austronesian languages,
lots of Papuan languages are like that.

A second-order letter monkey will produce two-letter
sequences in them with the  same relative frequencies
as the original it is aping. I remember  my colleague
Donald Laycock that, in some Papuan languages like
that, 80% of possible words existed in the language.
(A language such as Rotokas, with just 6 consonants
and five vowels, can ill afford not to make use
of every possible word-form). I vaguely suspect that
LSC sums would distinguish between real Rotokas and
second-order Monkey Rotokas. Third-order and beyond,
I am not so sure.

What do you think?

I am sure that, with a bit of effort, I can find
at least Luke or Matthews in Rotokas. But before
putting it  to the test, I must do more  thinking,
to emit a hypothesis that will be stand or fall.

Frogguy, with a thinking cap firmly glued on.