[kalwisti]

Originally Posted by AlexBell

Let me summarise my problems:

- The character in question is in Ancient Greek - the book is set in the 15th century
- I do not know any Greek at all, apart from a few characters used in high school maths
- The second Alpha in the word is shown as an Á in the Oxford Classic ebook version, and the Signet Classic print version; and I accept that this is incorrect
- I not have access to any example of the word shown in whatever the correct version may be.

So, can anyone tell me please what character/entity/glyph/markup I should use for the second Alpha to make the word look like the word carved on a wall of Notre-Dame de Paris in the 15th century?

Alex,
I believe that you have two options for solving this.

Option A:
Include the diacritical mark on the second Alpha. If you choose to do this, I would use Doitsu's solution -- given in post # 4:

The first capital Alpha would be encoded like this:

http://graphemica.com/%E1%BC%88
Greek capital letter Alpha with Psili (U+1F08)
< "psili" is the Greek word for 'smooth breathing [mark]' >
HTML Entity (Decimal):

PHP Code:

Ἀ 


The second capital Alpha would be encoded like this:

http://graphemica.com/%CE%86
Greek capital letter Alpha with Tonos (U+0386)
< "tonos" is the Greek word for 'tone, accent' >
HTML Entity (Decimal):

PHP Code:

Ά 


The diacritics are typically written at the upper left of capital letters.

Option B:
Omit all diacritics, except for the initial smooth breathing mark on the first Alpha. This was suggested by HarryT in post # 7.

If you choose this option, the first Alpha would be encoded the same as it is in Option A above.

(I prefer Option A, but since you're creating the e-book, you get to decide).

I understand your confusion; I think that although using the Greek-language grammatical terms (e.g., psili, tonos, oxia, etc.) is technically correct, it muddies the water for a non-linguist / non-specialist. The comments offered here have been very educational, and have opened my eyes to the challenges encountered when encoding something in a non-Roman alphabet.