2009/5/6 Eugene Oh <[log in to unmask]>

> My favourite is #1. Clearly I like to make life difficult for everyone. But
> still, in the native orthography I should expect something like this, where
> the speakers are expected to know—unless there's been a spelling reform or
> something. Even so, Japanese has had a spelling reform and still doesn't
> mark the accent.

Yet Modern Greek does (actually modern Greek corresponds to choice 5 here,
with the accent written down on every polysyllabic word, even if it could be
easily inferred from the form of the word). True, it has gone through a
spelling reform (the monotonic spelling), but even the polytonic spelling
from before the reform actually marked the stress on every polysyllabic
word. It just did so with various accents that originally marked something
else (Ancient Greek's pitch accent) .

This is just to show that the English solution is not necessarily the only
naturalistic one. Spanish and Modern Greek are just two languages that have
movable, unambiguously marked stress (although they resort to different
strategies to mark it).

> But that's just my mother-hen side coming out. Realistically and
> pragmatically, I also vote for #4. Efficiency. #2 takes up more space, and
> #3 reminds me too much of Irish orthography, which I find appealing, but
> monstrously difficult for non-native learners!

Personally, I like options 4 and 5, but if you are making a naturalistic
conlang (especially with a fictional), you should try and work out the
orthography not only in terms of practicality, but in terms of how it came
to evolve that way. So if it makes internal, conhistorical sense to mark
stress by doubling the vowel, by all means do so. If it makes more sense not
to mark it at all, just do so, clarity be damned.

On the other hand, if this orthography is just a transliteration tool rather
than the native orthography, I'd go for the maximum of clarity (since it's
an external tool rather than an integral part of the language) and choose
option 4 or 5.

You can even have a mixed solution. For instance, imagine that ai Basata was
unwritten until the area where it is spoken was colonised by the Spanish. A
written form of ai Basata would then clearly evolve influenced by the
Spanish model, including the rules of stress marking, especially since ai
Basata's stress patterns fit them relatively well. In that case, option 4
would be a given.

All in all, just think it through before taking a decision. It's not just a
matter of clarity. It's a matter of where the orthography comes from, and
what influenced it into taking this particular shape.
Christophe Grandsire-Koevoets.