Section on Background of Arabic Numerals

[behnam]

I've been doing extensive research about Arabic-Indic numerals and looks like we need to cover some of that background in ALReq. Here's a summary:

Common consensus in historic literature is that two variants of Arabic numerals were developed, both based on the Indic numeric system and Indic numerals. These two variants are usually called "Eastern Arabic Numerals" and "Western Arabic Numerals", and both are stabilized by the 800 C.E.

This chart demonstrates the common shapes for these two systems: https://en.wikipedia.org/wiki/File:Numeration-brahmi_fr.png

The Western Arabic Numerals—formed and used in Morocco and Spain, and used for centuries—were popularized by mathematics literature throughout Europe between 800 C.E. and 1200 C.E. and slowly evolved into European Numerals, which are encoded in Unicode as ASCII DIGITs. (See the same chart for other shapes formed during the evolution.)

The Eastern Arabic Numerals stayed stable in Eastern North Africa and Western Asia, which later evolved in two variations, which are encoded in Unicode as ARABIC-INDIC DIGITs and EXTENDED ARABIC-INDIC DIGITs.

IMHO, this history is very important in this context, because:

• it explains why, still to this date, the European numerals are the preferred form in Morocco and the Western Arab area in general.

• demonstrates that European numerals, as the modern shapes for Western Arabic numerals as also native to Arabic script.

Example:

What do you think?

Some References

The History of Mathematics, An Introduction, Burton http://www.maths.sci.ku.ac.th/suchai/02731141/hmath2.pdf

The Story of Numbers, UNESCO http://unesdoc.unesco.org/images/0009/000951/095141eo.pdf

Arabic Manuscripts: A Vademecum for Readers https://books.google.ca/books?id=NeaHnLb6RdUC&printsec=frontcover#v=onepage&q&f=false

Numerical Notation in Africa, Andrij Rovenchak (Ivan Franko National University of Lviv, Ukraine) https://www.afrikanistik-aegyptologie-online.de/archiv/2012/3553/

The History and Technique of Lettering, By Alexander Nesbitt

[behnam]

See also Origin of the numerals by Ahmed Boucenna, which proposes new roots for Western Arabic (Ghubar) numerals, also demoing examples in manuscripts.

[khaledhosny]

Seems a rather wild hypothesis based on the examination of single Arabic manuscript and no corresponding examination of Indian manuscripts.

[behnam]

@khaledhosny, that last paper on the roots of the system, yes.

But, "Western Arabic Numerals" are very well know and documented. I have tracked it back to A History of Mathematics (Florian Cajori, 1894). One problem is that these historical digits are not encoded yet, and that's why I'm writing a proposal for it at the moment.

Here, I'm not suggesting that we get into the details of the "Western Arabic Numerals", specially not getting into how these numerals look like exactly and how they were emerged/developed. But, there's clear consensus and enough evidence that European numerals are emerged from "Western Arabic Numerals", which explains many things and helps the audience understand the relation between the numerals and the scripts.

[khaledhosny]

Sure, Ghubar numbers (as they were indeed called back then) are different enough from modern numbers to be worthy encoding and the lineage between the two is well documented. I was commenting on the Arabic letters origin hypotheses, which does not seem to wild speculation not to different from the angles hypothesis.

[behnam]

From 1969 - Karl Menninger - Number Words and Number Symbols - A cultural History of Numbers:

[maurusian]

"Western Arab" is a misnomer. They should be called the "Moorish Numerals", since they were developed and used in Moorish Spain and Morocco. It's like an invention was made in Mexico by Mexicans, but it was called "the Spanish invention". It makes no sense.

[gounaman]

There are colonial as well as ethno/racist issues involved in the terms "Western Arab" and "Moorish". Might I suggest "Maghrebi Numerals" instead, or does that term also carry negative connotations for some?

[maurusian]

I think the word "Moorish" is the only one that both covers and specifies Andalusia and historical Morocco (including territories that are now part of neighboring countries such as Algeria and Mauretania) while being ethnically neutral. Using "Arab" means you excluded the contribution of the Amazigh (Berbers) and native Iberians. "Arabic speakers" or "Arabophones" may be more apt, but "Arabophone numerals" sounds contrived.

[shervinafshar]

@maurusian, @gounaman. I do appreciate your comments. If you'd be so kind to cite the literature your statements are based on, I will review the section and terminology in that light. Thank you.

[gounaman]

Re "Moor" as a Euro slur, there's this:

https://wp.nyu.edu/sjpearce/2016/10/31/not-a-moor-exactly/

Maghrebi (to me at least!) is more neutral, as it basically means from the western part (of the Arab word, that is) -- in a general, non exclusionary way. It encompasses, as a descriptive term, many ethnicities and nationalities in the North West of Africa.

But yes, it is an Arabic word, so maybe some might have an issue with that -- as the root of it is aghrab, or strangers, but also meaning The West, as in El Gharb.

To further muddy the linguistico-math waters, there's also this

https://www.al-fanarmedia.org/2020/03/the-latest-in-language-confusion-morocco-switches-back-from-arabic-to-french/

Sent with ProtonMail Secure Email.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ On Wednesday, June 23rd, 2021 at 10:20 AM, Shervin Afshar @.***> wrote:

@.(https://github.com/maurusian), @.(https://github.com/gounaman). I do appreciate your comments. If you'd be so kind to cite the literature your statements are based on, I will review the section and terminology in that light. Thank you.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[shervinafshar]

Okay. So our document is not using "Moor" at the moment and we are not planning to use it. The document is using "Western Arab regions" vs. "Eastern Arab regions" predominantly in geographical sense of the word. So the simile to Mexican inventing things that someone dubs them "Spanish Invention" is unclear to me. The document also is not recommending use of "Western Arabic Numbers" to refer to ASCII numeral. Having said that, I'm open to hear a more detailed clarification.

[gounaman]

So long as there is no potential confusion in the reader's mind as to the distinction between "Westernized Arab Regions" and "Western Arab Region". ;-) Jokes aside, it is awkward as a descriptive term; non-idiomatic, a little odd perhaps, to some. The appellation "North African" in English common parlance generally includes Tunis, Algeria, and Morocco, and with a bit of stretch, even Libya. see https://www.britannica.com/place/North-Africa Personally I like the term bilad as short hand for that part of the Arab world, as used by French-speaking North Africans. Also, this "Eastern Arab Region" nomenclature is a bit strange, too, As an Egyptian, I would take such a term to refer to the Gulf states or Oman. I certainly would hesitate to consider Egypt, Palestine, Syria, or Lebanon as located in the "Eastern Arab Region" -- maybe Jordan, on the outside -- wherever that "region" may be, exactly. Syria is called Bilad el Sham... so the term bilad does carries more of a universal authentic locale (not local) flavor, than the antiseptic regional alternative. Bilad El Arab is how the whole Arab world is generally known or referred to, or of course, 'El Umma' -- but that has connotations that you many not want to introduce in a technical document. Generally the more central part of the Arab world is traditionally referred to as the "Middle East," as we all know -- though that, too, has lingering, unpleasant English and French colonial associations. Yet, in Ar., we still see https://aawsat.com/ Sadly, even El Sharq El Awsat uses so-called "Arabic Numerals" in their Arabic language copy. Thus are cultures lost, if I may editorialize. I don't know if any of this helps, or just further confuses things, or simply sounds like Captain Obvious rattling on.

Sent with ProtonMail Secure Email.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ On Wednesday, June 23rd, 2021 at 5:09 PM, Shervin Afshar @.***> wrote:

Okay. So our document is not using "Moor" at the moment and we are not planning to use it. The document is using "Western Arab regions" vs. "Eastern Arab regions" predominantly in geographical terms. So the simile to Mexican inventing things that someone dubbing it "Spanish Invention" is unclear to me. The document also is not recommending use of "Western Arabic Numbers" to refer to ASCII numeral. Having said that, I'm open to hear a more detailed clarification.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[shervinafshar]

Thanks, @gounaman. I always enjoy talking shop about impacts of colonialism on naming things in the Middle East and North Africa, but if anything I'd like to narrow down this conversation on terminology and misnomers to a focused suggestion that could be cross-checked with (preferably) book sources and other experts and be applied to the document if needed. So far I've failed at that.

I invite all on the thread to review the terminology section of the document and keep the comments focused on what's reflected there and what are the editors recommendations and whether those recommendations are problematic. To summarize the document (and not the screenshots or sources mentioned in this issue) is using three distinct terms:

• ASCII Numerals
• Arabic-Indic Numerals
• Eastern Arabic-Indic Numerals

Personally I reckon that when talking in terms of Bilad El Arab or Umma, "Western" is anything to the west of Constantinople or Cairo and eastern anything east of those ancient cities.

[gounaman]

I read the terminology section last week. My comments here apply primarily to Arabic-Indic Numerals. Two immediate thoughts came to mind when I first read it: a) I would be more specific about the difference (or not) between the minus sign vs the negative sign, when it denotes a number that is < 0. Specifically, should the neg sum come before or after? Are there exceptions? What about the case with negative exponentiation? And does this need to be 100% or is it %100 consistent with the output of the JS International format object? And if Arabic numbers are read LTR, what is the impact of a trailing - sign, if any, from a BIDI/directionality perspective? Also, what about spacing considerations?

b) Fractions. I have an example (an Egyptian ad) where fractions can be represented as a numerator that is vertically atop the denominator, separated by a slightly tilted vinculum. Also the sequence of numerator/denominator cases must be nailed down with proof from primary source material. Wiki says otherwise from the terminology section. See: https://en.wikipedia.org/wiki/Eastern_Arabic_numerals#:~:text=Negative%20signs%20are%20written%20to,%D9%AB%D9%A1%D9%A4%D9%A1%D9%A5%D9%A9%D9%A2%D9%A6%D9%A5%D9%A3%D9%A5%D9%A8%E2%80%8E%20(3.14159265358). I may have other comments later on, as I am currently building an Arabic/ASCII calculator, using JS.

Sent with ProtonMail Secure Email.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ On Wednesday, June 23rd, 2021 at 6:44 PM, Shervin Afshar @.***> wrote:

Thanks, @.***(https://github.com/gounaman). I always enjoy talking shop about impacts of colonialism on naming things in the Middle East and North Africa, but if anything I'd like to narrow down this conversation on terminology and misnomers to a focused suggestion that could be cross-checked with (preferably) book sources and other experts and be applied to the document if needed. So far I've failed at that.

I invite all on the thread to review the terminology section of the document and keep the comments focused on what's reflected there and what are the editors recommendations and whether those recommendations are problematic. To summarize the document (and not the screenshots or sources mentioned in this issue) is using three distinct terms:

• ASCII Numerals
• Arabic-Indic Numerals
• Eastern Arabic-Indic Numerals

Personally I reckon that when talking in terms of Bilad El Arab or Umma, "Western" is anything to the west of Constantinople or Cairo and eastern anything east of those ancient cities.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.