Indesign CS4 ME - Diacritics in different color then letters?

Question

Hello again,
We have a lot of text in Arabic in our doc. Because we copied the text typed in Word 2007,  we lost the diacritic color set in Word (see pic). Is there any way to "turn it on" again?
Thank you very much in advance!

Answer 1

I had no idea I was going to write this much when I started... Sorry for the cultural-competence-in-typography rant, folks. I have to  chat with my favorite Arabic contractor this evening anyways, and he  loves it when I come up with excuses to spend a few hours on the  relations between Latin-script type and Arabic type after we've gotten  the mundane money-and-deadline stuff out of the way.
You know, my weakness here is that I don't know enough about the internal operations of the font to comment on whether or not the OpenType-driven composition of Arabic really reflects the calligraphic/typographic standards in Arabic. Harbs points out that
If the font uses GSUB tables to position the diactritics (BAD!), then
you are right. The diactritics will be part of a ligature, and can not
be styled separately.
If the diactiritics are positioned using GPOS
tables (CORRECT!), then they are totally separate glyphs and can be
styled independently
but at no point has anyone asserted that the OT methods actually reflect good typography in Arabic. For example, go read the Wikipedia article about Arabic diacritics, and note that the hamza section redirects you to a whole separate article about the hamza. There simply isn't anything in any European writing system that is analagous to it.
Wikipedia posted:
Although often a diacritic and not considered a letter of the alphabet,
the hamza (همزة), or glottal stop, often does stand as a
separate letter in writing, is written in unpointed texts, and is not
considered a tashkīl.
A "tashkil" is more-or-less equivalent to nikkud - it's a vowel for an abujida that doesn't have vowels. Is it a diacritic? Depends on who you ask.
The OP was asking for a feature (diacritic coloring) that is present in Word. If you take some Arabic text, turn on colored diacritics, and then switch from font to font, then you'll find that only the shadda gets highlighted in one sample when it's set in Times New Roman, but if you switch to Arabic Typesetting (feh) then the hamza on alef-with-hamza-below (or above) gets colored, even though it's part of a precomposed character.
So, I take this to mean that the ability of a given method of diacritic highlighting to actually highlight diacritics is going to be based on the construction of the font - and if the font isn't made by someone who is on top of Arabic typographic standards, then the identification of what is a diacritic and what is not is going to be misleading. So, I guess my message is: Don't assume that you can learn about Arabic type from Arabic fonts. The history of complex script typesetting is littered with bad decisions, crap workarounds, and implementations driven by monolingual nerds in the PNW whose comprehension of global culture basically boils down to knowing where to get good Thai food. </rant>
The Unicode spec, however, is a better choice for wrapping one's head around Arabic type, because it's been gone over with a fine-toothed comb by grumpy linguists and people who have been setting Arabic type since the days of hot metal. Of course, if you examine the spec, you'll find that not all of the separate "diacritics" are implemented in a given font, and not every "diacritic" in a font will necessarily have a Unicode ID (viz. Times New Roman). And, frankly, I don't know enough about Arabic to know which ranges of combining characters should be included in that GREP expression. For example, I think that including the hamza in the class of "diacritics" would be a mistake, but I want to check with my Arabic resources before I actually say that. But hey, as per Harbs & Jongware,
Anything that can be set using Diactritic Positioning, is fair game to
separate styling
and that's going to vary from font to font. Feh feh feh. So it depends on what the OP wants - coloration on anything that can be set with diacritic positioning? Or coloration on anything that should be changed, but can't in a particular font? The first is possible, but the second of course is not, for reasons that Jongware has already pointed out. But "in other words, you just have to try and see what happens" is a crappy method of controlling type rendering in a long document.
And PS: 1000x thanks, Harbs, I did get your email I'll be using that extensively once I put the Burmese to bed.