Re: Scripting was Re: Pablo is back, Job, Argentina, Relay, Lord of the Rings

On Tue, 15 Jan 2002 07:46:48 -0800
Peter Clark <pc451@...> wrote:

> --- Stephen Mulraney <ataltanie@...> wrote:
> > But what I really want is a 'combinatoric' script like JRRT's
> > tengwar - one which is based on a recurring graphical motif
> > e.g. stems and loops. But as much as I stretch my brains, I
> > can't quite imagine one that would look as nice as the tengwar. ;(
>
>    It's your decision, of course, but I would recommend against a
> combinatoric script, as it violates a key principle for clear and
> easy-to-read scripts: maximum distinction. As one wise sage once
> quipped, the elves must have suffered terribly from dyslexia. If you
> don't want a clear, distict script (this could be a legitimate choice),
> ignore the following.
Curiously enough, I find the tengwar rather readable - although I think
the principle  of maximum distinction is a valid one, I'm not sure if it
applies in the way you describe. One tends to get used to looking for
the particular contrasting part of a tengwa, while ignoring the common
elements such as the stems (apart from stem's height). Compare also those
arabic letters which are (in joined form) basically like a joined 'i' in
latin handwriting - but which are distinguished by dots above and below:
one dot below for baa', two above for taa', three above for thaa'.
Similar things apply to the separated forms, and to certain other sequences
pf letters: indeed 20 of the 29 letters are based on 9 shapes with varying
numbers of dots. Yet I find arabic script easier to read than, say cyrillic
which has a less 'unstructured' sequence of letters and to me is really very
eye-boggling. Of course, if one is designing a script with no 'combinatoric'
properties, the principle of maximum distinction would absolutely apply;
for a more structured script, one can design a repetitive 'substrate' such
as stems and bows, once the relevant contrasting parts do in fact contrast
easily.

Besides as you say, the eye tends to pick out word-shapes, not sequences of
letters - this suggests that a certain ambiguity in letter forms is acceptable.
This might be the mechanism by which I find it easier to see arabic than cyrillic
- there are less relevant contrastive elements since the principle of maximum
distinction is violated for certain (information-poor) parts of the scripts...?


>    - Scripting is an artform. The reason Tengwar looks nice is not
> because it is combinatoric script, but because Tolkien had a good
> aesthetic eye. If you don't have an artistic eye, the best way to
> develop one is to look at lots of different scripts. The internet is
> your friend--there are lots of sites out there that feature different
> scripts. DON'T look at individual letter forms. DO look at large bodies
> of texts.
Good idea - it's easy to become buried in the detail of individual letter
forms before you have a clear idea of the overall look of the script
which is what I'm aiming for. The idea is that an information-low
combinatoric basis will carry the aesthetic burden while other deviant
bits will carry the information ;). Perverse, but the idea intruiges me ;)


>    <rant>I personally find Cyrillic rather difficult at times,
> especially if the typesetting is poor. The letters for sh, shch, ts, i,
> p, and n can all run together if placed too close to each other. And
> don't get me started on handwriting! I forget which, but there is a
> Russian word that when written cursively is little more that "UUUUUUUU"
> or something like that (as best as I can represent handwriting in ASCII
> text).</rant>
A word made completely of the letters geh, yeh, ee, ee kratkoye, ell, em,
en, err, teh, tseh, shah, and shchah, then; yep that's a lots of letters
that look like just wiggles in handwriting ;)


> > It's very distracting; I really should be learning quantum mechanics
> > and looking for a job (unrelated activities, I should say...)
>
>    Something more important than conlanging? Nonsense. Sir, we will
> have to ask you to leave this list immediately. And take your heretical
> ideas with you! :)

Well, note the use of the word 'should': it's not happening ;)

>    Seriously, nothing is unrelated to conlanging, because conlanging is
> language, and if you can't express something something, or even a part
> of something, then you've discovered something very rare and quite
> special.
;) Actually, I meant that QM is unrelated to getting a job, which probably
goes without saying...

Thanks for your interesting mail, Peter!

Stephen

Reply