hamza

Wow... good comments up to this point. My question is on the sole citation of William Wright up to this point, in the beginning where the origin of is implicitly or briefly discussed. Due to essentially an overflow from a debate on the discussion page of Middle Bronze Age alphabets, this has kind of become relevant. I am not suggesting that a templated 'origins' page be added to each letter, as this would create significant controversy in determining single creation myths for each letter.

I am wondering two things (if people strongly disagree please revert this) a) is William Wright the best source - particularly since better sources are available - his work is more than a hundred years old and there has been some scholarship in the meantime (which is at the bottom of the article, granted); b) the reference is misleading. On page 15, in the orthography of hamza, Wright ([1]) describes in Rem B that sometimes the in conjunction with alif produces a sound like an 3yin, particularly in African dialects. Frankly I cannot follow a lot of his archaic abbreviations in English (or I'm stupid). But my take is that one particular case sounds like an 3yin and was written as an 3yin in some dialects. That has no bearing on the actual evolution of - only on the interaction of pronunciation and writing in certain dialects (mind you not in Classical Arabic it would seem according to Wright).

According to Islamic Awareness (http://www.islamic-awareness.org/Quran/Text/Scribal/haleem.html), which I have found to have very good historical Arabic information available, including very good collections of photo of Arabic inscriptions, these three men (Abū-l-Aswad al-Du'alī (d. 69 / 688), Nasr Ibn `Asim (d. 89 / 707) and Yahyā Ibn Ya`mur (d.129 /746)) introduced yellow dots for and red dots for diacritic marks. The implication is that the may not have existed as a distinct character in Arabic, but rather only as part of a split pronunciation of alif (as either modern or modern alif). This idea is echoed in a few other sources of Arabic linguistics. For the time being, I am going to remove this claim, particularly since it is not relevant to Classical Arabic even in Wright's book. Michael Sheflin (talk) 06:02, 25 September 2009 (UTC)

Unicode has separate glyphs for plain and "wavy" hamza, eg. U+0625 ARABIC LETTER ALEF WITH HAMZA BELOW and U+0673 ARABIC LETTER ALEF WITH WAVY HAMZA BELOW. Could somebody explain the difference and add it to the article? Jpatokal (talk) 04:43, 28 June 2012 (UTC)

The wavy is used in Kashmiri for long /ɨː~əː/.--Lüboslóv Yęzýkin (talk) 07:42, 24 November 2015 (UTC)

The yaa harf should not have two dots under, when it is in the beginning or in the middle? — Preceding unsigned comment added by 85.244.145.70 (talk) 12:30, 17 November 2012 (UTC)

When the bearer is a yaa, it should always be written without any dots regardless of its position in the word. When a yaa does not bear a it MUST have two dots under it in its initial and medial shapes; whether the dots are also to be written under the isolated and final shapes varies according to style: AFAIK in that case the dots are usually written in the Maghreb and not written in the Mashreq, hence the Unicode name of "Farsi yaa" for a yaa which has dots in initial and medial shapes but not in isolated and final shapes. — Tonymec (talk) 13:03, 15 September 2015 (UTC)

I suggest you to look through the book Mu'ajjam tasreef al-af'aal al'arabiyya by as-Safeer Antoine al-Dahdaah, there are examples of every type of hamzated verbs including hamzated and doubled, hamzated and middle-weak (with madda) etc. When I used to study the hamza-rules it helped me very much. — Preceding unsigned comment added by Mezei nelli (talk • contribs) 12:02, 18 March 2014 (UTC)

I've heard still another description; I think it was in "Teach yourself Arabic":

In the Prophet's time and place of birth, the had ceased to be pronounced, so the first Qur`ān texts came to be written with no hamzahs, except as word-initial alephs. Later, grammarians reestablished the hamzah but they wouldn't alter the holy text. So: first write the text as you would if no hamzahs were to be pronounced, then add the hamzahs while removing any dots under hamza-bearing yaa letters. — Tonymec (talk) 13:14, 15 September 2015 (UTC)

It is thus. Hamza has been elided in the dialects of the tribes of Quraysh and Tamim and in Hijaz generally at a very early stage (pre-Islamic). Under the influence of the nearest i or u became y and w respectively, hence its current orthography with ya and waw. Later it was reintroduced from (supposedly) Bedouin dialects. When it was lost altogether (usualy word-fanally or after consonants or long vowels) the independent on the line was required and hence was invented. To sum up, if one knows this little pre-history the spelling rules for become not so difficult.--Lüboslóv Yęzýkin (talk) 08:25, 24 November 2015 (UTC)
Some sources which explain: [2] [3]. I suppose it might be added somewhere.--Lüboslóv Yęzýkin (talk) 08:34, 24 November 2015 (UTC)

Can we stick with one spelling, either or hamzah, throughout this article? - dcljr (talk) 05:32, 16 October 2015 (UTC)

Done ("hamza", in accordance with the Unicode name of the character). -- Karl432 (talk) 12:44, 16 October 2015 (UTC)

what do the different colors in the tables mean? Backinstadiums (talk) 15:36, 26 November 2016 (UTC)

I've added an explanation.--Lüboslóv Yęzýkin (talk) 10:08, 29 November 2016 (UTC)

a) ط is a grapheme which can be connected both to the previous and to the latter letter, so it does not work for some rules. b) Which combinations are not attested in MSA?Backinstadiums (talk) 15:46, 26 November 2016 (UTC)

@Backinstadiums:
A) It was done on purpoese. The ط is a filler symbol (I tried a square but it looked ugly). The goal is not to show every shape of a letter but the spelling of the in various contexts.
B) For that we must analyse an Arabic text corpus. I did not stumble upon such researches. So the table shows only potential combinations and their graphic representations according to the spelling rules.--Lüboslóv Yęzýkin (talk) 09:46, 29 November 2016 (UTC)

@Любослов Езыкин: Here's a word version of Lane's : https://lanelexicon.com/updates/Backinstadiums (talk) 10:26, 29 November 2016 (UTC)
@Backinstadiums: I meant already existed scientific analyses published in journals or somewhere, so we could cite them as reliable sources. Of course, there exist corpuses and wordlists in Arabic. But, if you really need an answer to your question, you can analyse yourself whatever the corpus you choose. But I myself refrain from doing that and I do not think it that important for our article. It's more about corpus linguistics. --Lüboslóv Yęzýkin (talk) 10:53, 29 November 2016 (UTC)

I am starting to learn arabic, so I do not come up easily with examples. I know using regular expressions, regex, and a corpora or dictionary it could be done. Yet, I do not master regex either, let alone in arabic script, so any help would be appreciated. Thanks in advance. Backinstadiums (talk) 21:14, 16 January 2017 (UTC)

I understand that there're different traditions regarding the spelling rules of hamza. If possible, please specify which one this page shows, as well as the alternatives ones to add them. Lastly, a grammatical/academic reference should be mentioned. Thanks in advance. Backinstadiums (talk) 21:19, 16 January 2017 (UTC)

A link to the page of the rules wikipedia/wiktionary follow should be added. I cannot find it anywhere. Thanks in advance. Backinstadiums (talk) 21:42, 16 January 2017 (UTC)

@Backinstadiums:I don't know of any Wikipedia policy but wiktionary uses attested spellings rather than trying to standardize on any rules, and includes all attested spellings with redirects etc. — Radix^cc 📞 02:24, 26 July 2017 (UTC)

@Любослов Езыкин: The current rules don't match up with with some spellings I'm finding such as شؤون. I'm wondering if it's ok to change the page to reflect the rules in A Reference Grammar of Modern Standard Arabic by Karen C Ryding because I have a reference for them and they seem to match spellings I find. Differences:

1. Initial Hamza is the same.
2. Medial Hamza seat is determined by vowels on both sides whether long or short with the usual i u a priority. Diphthongs y w are treated like i u.
3. Medial lone Hamza on the line happens only for ā'a ā'ā or any case after ū.
4. Final Hamza is lone on the line after a long vowel or consonant ending and has a corresponding seat after a short vowel ending.

Radix^cc 📞 02:24, 26 July 2017 (UTC)

@Radixcc: I'm OK with anything, as my only contribution to the article was the overview table. I do not remember what sources I used, they must be several, but my baseline source must have been Faruk Abu-Chakra's Essential Grammar (chapters 7 and 20). I do not know why or where Abu-Chakra and Ryding may disagree, of course, if they disagree. I was concerned about this matter a couple of years ago and I haven't studied it since then, so I must forget something and I may not remember/know every of its intricacies. I'm not sure why I thought that uʾū is ـُءُو and not ـُؤُو.--Lüboslóv Yęzýkin (talk) 18:00, 27 July 2017 (UTC)
Maybe I followed the rule: "avoid two consecutive wāws"?..--Lüboslóv Yęzýkin (talk) 18:10, 27 July 2017 (UTC)
I used the Quran as my text for testing, and it writes رءوس for رؤوس. Archaic spelling?--Lüboslóv Yęzýkin (talk) 18:35, 27 July 2017 (UTC)
See also Caspari & Wright for this particular case.--Lüboslóv Yęzýkin (talk) 19:10, 27 July 2017 (UTC)
@Любослов Езыкин: Ok, I got the Abu-Chakra book and his rules seem significantly more complex. Ryding's book was sponsored by an Arab government so might represent a modern state-sponsored simplification effort even though it was published two years before Abu-Chakra. I think I'll just include Ryding's simplified rules and note the cases where Abu-Chakra differs. Regarding the Quran, I don't think it can be used as a good example of modern spelling since some words like كتاب are spelled much different like كتٰب so could be equally or more unusual compared to the modern. Though it's the most common example of fully vocalized Arabic so is tempting to use as a source of examples, it has quite a few differences with modern language use. — Radix^cc 📞 23:08, 27 July 2017 (UTC)
@Radixcc: 1) I doubt that Abu-Chakra's rules are more "complex", they seem to be just the same, only worded differently and in a more formal and obscure way. Rydings rules are more simply and straightforwardly formulated. There is also other similarly simply formulated rules from some other old grammar. Yet they all fail to mention more complex cases with two wāws like رءوس vs. رؤوس.
2) Yes, there are Arabic corpora in the web, but few as well-made and easy-to-use as the BNC, so the Quran looks like the most accessible ad hoc corpus for a quick check.--Lüboslóv Yęzýkin (talk) 13:09, 28 July 2017 (UTC)

There are a few inconsistencies between the description and the tables. Probably it'd be best to stick to one set of rules and use it in both the description and the tables or to sketch the differences in the set of rules laid out by different authors. Example of the inconsistencies: according the detailed description, medial preceded by ū and followed by i should sit on a yā’, but the table shows hamzah written on the line. Ago (talk) 17:36, 14 June 2019 (UTC)

I was going to comment the same thing. These are all the inconsistencies I noticed:
• Under "If the is final":
  • "If a short vowel precedes, the is written over the letter (alif, wāw, or yāʾ) corresponding to the short vowel."
    • The table follows this rule, but also lists alternative forms with the written on the line (e.g. both طِئ and طِء‎ are listed for ṭiʾ).
• Under "If the is medial":
  • "If a long vowel or diphthong precedes, the seat of the is determined mostly by what follows: If i or u follows, the is written over yāʾ or wāw, accordingly. Otherwise, the would be written on the line. If a yāʾ precedes, however, that would conflict with the stroke joining the yāʾ to the following letter, so the is written over yāʾ. (as in جِئَت)"
    • In the table, a following ī or ay is always written over yāʾ, a following ū is always written on the line, and sequences containing a following aw are listed with two spellings, one with on the line and one with over wāw, except for ṭawʾūṭ which only has the spelling that's inconsistent with the rule (طَوْءُوط‎ instead of طَوْؤُوط‎). For example, ṭīʾuṭ is spelled طِيئُط‎ in the table, when according to this rule it should be طِيؤُط‎, and ṭūʾiṭ is spelled طُوءِط in the table, when according to this rule it should be طُوئِط. Additionally, ṭāʾūṭ is spelled طَاءُوط‎ in the table, when according to this rule it should be طَاؤُوط‎.
    • The example given for the last sentence makes it seem like a preceded by yāʾ and not followed by i or u should be written on the preceding yāʾ and not on a separate yāʾ, since the example only has a single yāʾ, but the table such sequences are spelled with the written on a separate yāʾ, e.g. ṭīʾaṭ is spelled طِيئَط‎ rather than طِئَط‎ (in this case I suspect that the example in the description is misleading).
  • "Otherwise, both preceding and following vowels have an effect on the hamza. If there is only one vowel (or two of the same kind), that vowel determines the seat (alif, wāw, or yāʾ)."
    • According to this rule, ṭūʾuṭ, ṭūʾūṭ, ṭāʾaṭ, and ṭāʾāṭ should be spelled طُوؤُط‎, طُوؤُوط‎, طَاأَط‎, and طَاآط‎, but in the table they are spelled with the on the line. Similarly, ṭʾūṭ should be spelled طْؤُوط‎ according to this rule, but in the table it is spelled طْءُوط‎.
The description is also incomplete, since it does not describe what happens when an initial is followed by ī or ū. The tables are incomplete as well, since they do not cover the case where a final is preceded by a diphthong or a consonant, nor do they cover the case of a preceded by a diphthong and followed by a consonant. Zgialor (talk) 19:47, 13 January 2021 (UTC)

The "High Hamza" section needs work, but I'm not sure how to improve it.

The table for "high hamza" in Hamza#Orthography using the {{Arabic alphabet shapes}} template comes up as blank for me, the cells all say "(none)", which doesn't seem very useful.

Image of what i see:

The associated text doesn't make much sense either:

By itself, as a high Hamza (not used in Arabic language; only one isolated form, but actually used in medial and final positions where it will be non joining), after any Arabic letter (if that letter has an initial or medial form, these forms will be changed to isolated or final forms respectively):

I'm not sure how to clarify it, because i'm not sure what it's trying to say, so I've left it alone for now.

Irtapil (talk) 20:47, 25 September 2020 (UTC)

That description is phrased in a confusing way, but as I understand it, it is simply stating that high (1) has only isolated shape, (2) is used in medial and final positions, (3) characters around it need to be shaped not to join it (duh). The (1) is true, (3) is a trivial consequence of (1), but I would like to see the source for the (2). Looks like in Kazakh it's used only at the beginning of words. Is it used in any other language?
On your screenshot, the cell isn't entirely blank but contains about two barely visible dark pixels. This seems to be a problem with fonts. For me it renders perfectly fine in Webkit (slightly off-centered within the cell if you look closely) and truncated roughly in half in Firefox. Looks like the problem might be in that some fonts incorrectly treat U+0674 as a non-spacing mark. In Scheherazade, which is preferred by the template, it has a zero advance width. How about this:

Isolated	Isolated	Isolated
ٴ‎	ٴ ‎	ءٴء‎

Does this help? If it does, then it might be possible to work around the problem within the {{Arabic alphabet shapes}} template. For me, the middle one doesn't get truncated in Firefox.–MwGamera (talk) 00:29, 27 September 2020 (UTC)

Some versions of the Quran have the "regular" + "light" represented by a filled black dot over the single alif. Is there any specific name for this letter combination? Is there a UNICODE symbol for this? Francois (talk) 08:21, 23 August 2021 (UTC)

You mean this, right?

 

I have no idea if it's appropriate but it looks like it's often spelled in Unicode as ءَا۬عْجَمِىٌّ using U+06EC ◌۬‎ ARABIC ROUNDED HIGH STOP WITH FILLED CENTRE. –MwGamera (talk) 10:54, 23 August 2021 (UTC)

Yes, that's it. Thank you. Francois (talk) 08:23, 28 August 2021 (UTC)

I think the page should indicate ways that Quranic orthography is different in representing hamza, largely because the rasm of the Quran was standardized without the representations of hamza, which were later added as diacritics and could not be added on top of a seat that was not already written. The differences are mostly:

• آ (alif madda) is not used to represent /ʔaː/; instead ءَا is used. This is because madda is instead used over ا, ي, و to mark extra-long vowels that occur in some environments in Quranic recitation; the madda was not used for /ʔaː/ until later, I believe. For example, Quran 54:41 has ءَالَ for modern آل.
• ئ is written like ىٕ (I used alif maqsura with combining below) when it is followed by a kasra. For example, Quran 2:5 starts with أُو۟لـٰۤىِٕكَ for modern أولئك.
• ؤ, ئ, ىٕ are not used if they would lead to two waws or two ya's in a row. Instead, ء is used. For example, Quran 18:31 has مُتَّكِـِٔينَ for modern متكئين
• If the letter before a that would normally be on the line on the line would connect to the letter after the hamza, is written above the connection instead of on the line, like ـٔ. You can see this too in مُتَّكِـِٔينَ

I'm not really sure I can add this myself. I'm not sure where to find a reliable source, but by googling I can find religious websites that provide rules like this, for example here [4] in the section called III. The Hamza. Dijekjapen (talk) 05:20, 25 January 2023 (UTC)