On 04/03/2011 13:49, Philip Newton wrote:
 > In it, the author describes how _šnu_ "sing" in Kabyle
 > apparently derived from Arabic _šani'a شنئ_ "hate"...
 > which seemed a bit improbable at first, until he found a
 > plausible link. (Read the article for details.)


> Anyway, I thought it's an interesting article on
> conlanging, especially in the case of diachronic
> development or borrowing, where derivation and/or cognacy
> is obscured by shifts like these.


> And that if all
> relationships between words are obvious, the development
> may not be very naturalistic :) (For example, if a
> Romlang can be easily read, with correct comprehension,
> by anyone with a good knowledge of Vulgar Latin and a
> cheat-sheet of sound changes.)

Yes, indeed. Equipped with a cheat list of sound and
spelling changes from Vulgar Latin to modern contemporary
French, for examples, may cope with a quite a bit of
vocabulary, but there are an awful lot of exceptions and
lacunae. Two obvious ones come to mind:

One would expect VL _vespa_ (wasp) to be *vêpe - it ain't;
it's _guêpe_. Blame Germanic settlers for that.

One would expect VL _a'more_ (love) to give *ameur - it
doesn't (tho i have been told _ameur_ survives in dialect
for the rutting of animals); the actual modern French word
for 'love' is _amour_ - an irregularity? Well, yes - it's
borrowed from southern Occitan. Probably partly because the
influence of the troubadours but also, I suspect, because the
southern word sounded more 'pink & fluffy' than the regular

Applying the great sound change/ orthography change to verb
forms would produce nothing much resembling the modern
French verb system where analogy, leveling out of
irregularities etc has been rife.

There's also change in semantic. Altho _poison_ is derived 
regularly from VL _poti'one_ (drink), it doesn't mean 
"drink" in modern French!

And both French _blesser_ (to wound) and English _bless_ are 
from the same German root, the modern words don't mean the 
same. And why did the French abandon the VL _vUlnE'rare_ in 
favor of this Germanic word?

And so one could go one and on and on.

If one wants a Romlang to appear naturalistic, there's a
good deal more to be done than just applying the "master
plan" to Vulgar Latin  ;)

The same considerations, of course, would apply to a
Slaviconlang, Celticonlang or Germaniconlang etc.

Frustra fit per plura quod potest
fieri per pauciora.
[William of Ockham]