[from RW:]
/For metrical purposes/, we don't know whether the syllable is open or
closed until we know what comes next. [emphasis added]
About that you are right, and it was an oversight on both our parts. But
the dictionary also contains πράσσω with ᾱ in an annotation, and the
weight of the syllable containing the ά is unambiguous (heavy).
Given the nature of the actual annotations (often with examples as
proofs), some omissions are probably just because they didn't know. And
as for typographic matters, it seems they would have used diacritics
above ᾱ, ῑ, ῡ, had the relevant pieces of metal type been available.
Unless they were trying to make the point that it must be an annotation
because we sometimes don't know – but then the macron wouldn't appear in
some headwords in some positions.
[A] document [...] using macrons and breves as part of the text [i]s
www.aoidoi.org/articles/meter/intro.pdf . It writes them above the
tone accents!
If you mark meter for scansion, you do it for every syllable, and then
it's of course above the entire text string; btw the placement in that
paper (eg: p. 5) is clunky (I'd put the marks all at equal height),
which screams "higher-level protocol" :-) Scansion is a slightly
different use case for the breve and macron.
The Oxford Grammar of Classical Greek states explicitly (p. xii) that it
uses the macron to indicate that a /vowel/ is long, and it contains
examples such as:
πρᾱ́γματι (p. 6)
κῑ́ων (p. 215)
πρᾱ́ττω (p. 228)
all with the oxia/acute above the macron, like in the Pocket Oxford
Classical Greek Dictionary (but author and publisher are the same). I
suppose you can argue about a breathing mark being on a separate tier,
but semantically, a vowel length mark binds more tightly to the letter
than an accent.
[from PV:]
they just should be documented somewhere for implementers of renderers
and fonts, to support these types of clusters
Now this is something I wholeheartedly agree with. For many languages
and orthographies.
Stephan
