don't confuse!

Perception is multilayered.
Some "see" blue, black and white.
others see strokes and dots,
others Arab words, or words from the beginning of the qurʾān or (among other things) 18 "dagger alifs".     I see only nine.
A "dagger alif" is a small alif, an Ersatzalif (supple­mentary alif)
or a Wandelalif (transforming alif).
The similar looking signs in the "Indian" muṣḥaf from Medina
are no alifs, but fatḥas, standing fatḥas or turned fatḥas,
not letters but vowel signs.
I fail to understand, how anybody can confuse a sign that sits on its letter (or near its ascender)
with a letter that follows a letter+vowel­sign-com­bina­tion.
Once one knows that in the African notation there must be a ḥarf al-madd to lengthen the vowel,
where­as in the Asian notation fatḥa, kasra, ḍamma each have a turned vari­ant for the long vowel,
one SEEs the difference.
Whoever does confuse long vowel sign and Ersatz letter is blind.
In the last line: two Wandel­alifs:
there is a ḥarf al-madd, but the wrong one: instead of alif (expected after fatḥa): waw resp. yāʾ here the small alif transforms waw resp. yāʾ into alif. (In the blue line above: (long) standing fatḥa again.)
So not 18 dagger alifs, just seven Ersatzalif plus two Wandelalif.
Not convinced?

Look at these examples.
If you don't see "argu­ments", study some­thing else (you are not made to study the writing of maṣā­ḥif).

Here the text from the King Fuad Edition about the small letters (among them "dagger alif"):

And about fatḥa, standing fatḥa, kasra, standing kasra, ḍamma and turned ḍamma.

Qurʾānic Orthography

Most people think that there is ONE way to write the qurʾān, that if we leave the different qiraʾāt aside and ignore the style of writing, all copies of the qurʾān are the same.
And that they are written alright.

The opposite is true:
there are many different qurʾanic "ortho"graphies and none is ortho/right.

b, k, l, m, n, t ... are fine. Paleo-Linguists argue about ض ص ط ظ

But for modern readers only the vowels are proble­matic: short and long vowels.

Basically there are two systems:
an Indian system with seven vowel signs (a ā i ī u ū x)
an African System which needs for each long vowel a vowel sign plus the correspond­ing lengthening letter.

The African system is today common in the Arab world. When there is no leng­then­ing letter (waw, yāʾ, alif) after ḍamma, kasra, fatḥa in the rasm, a small letter is added. When the rules of pro­sody require an "ī" al­though no yāʾ is in the rasm, a small yāʾ barī (i.e with the tail to the front) is added.

But when the rules of prosody require a (written/long) yāʾ to be shortened,
that is not reflected in the text.
I am shocked because the lengthening of vowels required by prosody IS shown.
I had no problem with sticking to the base letters, but adding read­ing helps sometimes defies God's logic.
(Turks and Persians do never show the niceties of quranic as­simi­lation. ‒ That I can under­stand.)
(Turks note leng­thened ī, but not leng­thened ū ‒ some­thing corrected in all Indo­nesian and some Iranian reprints.)

Indians, Turks, Persians, Indo­nesi­ans are not happy with this. The Arab attitude "every­body knows that these letters do not leng­then the vowel at these places" sounds arrogant in non-Arab hears.

In XXX:10 /ʾasāʾŭ s-sūʾā/ in India (second line) and Indonesia (last, right) there is no vowell sign about the wau in /ʾasāʾŭ/, hence it is not pronounced (just as the following alif).
In Kerala (first, right) and Turkey (fourth line) that wau is silent because it is only the "seat" of hamza (there would be written "madd" under­neath if it were to be pronounced).
In the new Iranian orthography (fifth line, right) all silent letter are pink.
But in the third line (right: Madina: ʿUṯmān Ṭaha; left: Damascus: Dār al-Maʿrifa) there is no Silent-Sign above the wau.
The same in the last line, left (Tunis: Nous-mêmes) ‒ wrong as I see it.
But two of the taǧwīd-Editions based on UT do show the silence of that wau: On the top, left from Bairut (blueish for silence), on the fifth line, left from Damascus (grey triange above for silence).

When shortening doesn't follow a general rule, but applies just to parti­cular places within the text = when a vowel is short because it must rhyme with lines before and after, this IS reflected in the Arab-African text. So why not: all the time?

Here you see words from an Indian manuscript (from Surat Hūd) in which ONLY the vowel SIGNS count, the "lengthening" vowel letters are IGNORED (hence: NO sign above or below) ‒ al-farīqaini in the last line has jazm above the yāʾ because in the diph­thong yāʾ is NOT silent.
((Added later: I first had seen just one manu­script from around 1800 with the sign-only ortho­graphy, by now I have seen some more ‒ up to the time when litho­graphs became common, litho­graphs with the modern/ mixed way of writing long vowel letters)

on the margin I added doctored versions: signs were moved to a place easier to read for the modern reader

The modern Indian system (black on white background) is a mix of the old con­sistent system and the African one: when the CORRES­PONDING letter follows a vowel sign, the SHORT vowel sign is used (as in Africa), only when there is a diffe­rent letter or no "leng­thening" letter at all, the long vowel sign is used.


In 7:103, 10:75, 11:97 and 43:46 (and 10:83 with an added mīm for plural) pro­nounciation and rasm are the same; there is only disagreement on whether the alif or the yāʾ is mute:
wa-malaʾihī
IPak: وَمَلَا۠ئِهٖ
Q52: وَمَلَإِي۠هِۦ
In the rasm there are matres (ḥurūf al-madd) for both /a/ and /i/,
‒ indeed here letters (not consonants!) stand for short vowels,
  because there was no other way to notice them.
in India alif is silent (the short fatḥa is valid, not the alif), yāʾ carries hamza,
in Arabia alif "carries" hamza below, yāʾ is silent.

In 21:34 ʾa-faʾin
IPak: افَا۠ئِنْ
Q52: اَفإي۠ن
India and osm/Tur make the alif silent
(Indians used to leave the alif without any sign, now they put the silent making circle,
Turks write qaṣr underneath)
for Arabs alif carries hamza, yāʾ is silent.
Muṣṭafā Naẓīf in one of his manu­scripts (the 604 pages berkenar one) just drops the otiose letter. اَفَإنْ
It does not help to observe that in his other maṣāḥif he has the super­fluous letter. The 604 page muṣḥaf is often re­printed (not as often as the 522 page one, but in different countries) without "correc­tion".
Similar 6:24 min-nabaʾi

In Q52 alif "carries" hamza and /i/, yāʾ is silent and in IPak alif is silent, yāʾ carries hamza and /i/.