--- On Mon, 6/6/11, Michael Everson <[log in to unmask]> wrote:

> > Interesting. I don't see the "order" itself as an
> > intrinsic property of an alphabet.
> 
> The history of alphabets suggests that it is. Latin, Greek,
> Cyrillic, Armenian, Georgian, and a host of other alphabets
> do share the basic Western Semitic order.  Where that
> has changed (as for Brahmi, subsequent daughter scripts
> follow the new order. 
> 
> > As far as I'm concerned, ABCDEFG and CPENDLS (plus the
> remaining 19 letters) *are* the English alphabet. The
> English alphabet is a set of 26 particular letters.
> 
> I'd disagree. The order is intrinsic.

Fair enough. Suffice to say, opinions differ!

> > If you view order as an intrinsic property to an
> alphabet, then I would say that "CPENDLS" a) can NOT be the
> English alphabet
> 
> It can, if for a particular alphabet a reform has been
> instituted.

As far as I'm aware, we haven't had such a reform! The fact that something
as insubstantial as a "reform" can allow this new order to be identified
as the "English alphabet" indicates to me even more that the specific
order of letters is not a property of at least this alphabet. Others may
vary.

> In the absence of such reform however, the
> inherited order tends to be kept. Even if slightly tweaked
> here or there. 

Tweakage also indicates that order is not an intrinsic property.

If we changed this to musical scales, "tweaking" the C major scale by
adding or subtracting a couple notes or changing the order of notes leaves 
us with something other than the C major scale. This is because note order 
is an intrinsic property of a scale (as opposed to a "key", for which note 
order is not relevant).

A "reform" could not, for example, alter the nature of a musical scale.
It has certain note intervals and is not changeable. If you changed it,
you'd end up with a different scale with different properties.

> > and b) "begins with AB" and "ends with Z" are equally
> fundamental to the nature of the English alphabet. Because,
> if you put the Z anywhere else, you no longer
> > have the intrinsic property of particular order.
> 
> Tailoring does not necessarily invalidate the intrinsic
> order, so long as the skeleton is kept. 

I could argue that tailoring does invalidate the intrinsic order.

Or that such tailoring simply indicates that order is not an intrinsic
property of this alphabet.

> That is, Icelandic A
> Á B C D Ð E É F G is just as Latin as Irish A(Á) B C D
> E(É) F G is. 

Well, one is Icelandic and the other is Irish. Both alphabets are 
certainly derived from the Latin, though. I don't think many Roman
writers would know what to do with accented letters or runic letters!

Padraic