Talk:Arabic numerals

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Medieval Arabic numbers[edit]

We can see the correct format and sequence of the medieval "european" numbers in title page of the book Libro Intitulado Arithmetica Practica by Juan de Yciar, the Basque calligrapher and mathematician, Saragossa 1549, and at Filippo Calandri, De Arithmetica, Florença: Lorenzo Morgiani and Johannes Petri, 1491-92, page 145. The figures shows the calligraphical place value of: one 1, two 2, three 3, fuor 4, five 5, six 6, seven 7, eight 8, nine 9, and ten o.


Don't Arabs today write numbers left-to-right?![edit]

"From the point of view of the reader, numerals in Western texts are written with the highest power of the base first whereas numerals in Arabic texts are written with the lowest power of the base first." Is this true today? I don't read Arabic, but I think I've noticed in Arab newspapers and Arab TV that their numbers - like everybody else's - run left-to-right? This should be able to confirm and then correct the article. - Ben Franklin (talk) 15:49, 10 May 2013 (UTC)

Arabic is written right-to-left, but, as the article says, Arabs write the digit with the lowest place value first. For example, writing "five hundred and forty-three" would be done in this order:
Step Western Arab
1 5 3
2 54 43
3 543 543
For someone reading English (left-to-right), the digit 5 comes first, but for someone reading Arabic (right-to-left), 3 comes first. --Joshua Issac (talk) 16:11, 13 May 2013 (UTC)
Let'a put it another way. When you were a little one in school, did you ever wonder why we WRITE from left to tight, but DO OUR "SUMS" from right to left? (Just picture a column of figures, and well, go figure!). I actually asked my old Mum when I was five, and being a clever old thing she gave me more or less the right answer! If we'd invented our numbers off our own bat instead of copying them off the Arabs we'd have all our account books set up to go from left to right (well, wouldn't we!!!) so its US that go "the other way" when it come to numbers! (Although to be fair the Arabs, clever things, copied them off Indian mathematicians originally). --Soundofmusicals (talk) 02:58, 19 May 2013 (UTC)

When you see how Arabic numerals are spoken aloud, you'll find that numbers between ten and ninety-nine are spoken with the general pattern to first say the word for the units digit, followed by the word for the tens digit (which is generally the word for the digit with -'un added). As the written words follow the ordering of spoken words, Arabic writing of digits continue in the general left-to-right direction as text. See for example: [1] There are some irregularities to the general pattern, as the Arabic words for 10 and 20 are a little different from 1-'un and 2-'un, but the general pattern holds form. By comparison, English has special words for 11 and 12, then speaks 13 to 19 in a manner following the left-to-right Arabic pattern, but completely switches direction to right-to-left for 21 through 99, where the tens digit is spoken first, with -ty added, followed by the ones digit. So 24 is spoken right-to-left as "twenty-four," while for English to follow the left-to-right pattern, we'd be speaking 24 as "four and twenty," which I've only seen in nursery rhyme. [2] Cafehunk (talk) 00:43, 17 March 2018 (UTC)

The fact that in Arabic the numbers are spoken lowest-digit first is excellent information and should probably be inserted into the text about use of these digits in Arabic right-to-left text, is there any kind of reference for it? I'm guessing an example would be that 25 would be spoken as 5,2-'un.Spitzak (talk) 18:04, 10 April 2019 (UTC)


There is a difference between Arabic numerals and Hindo one; the Arabic ones are those used in European languages, while the hindo are those still used in Arabic language itself like ٠١٢٣٤٥٦٧٨٩.

الرشيد (talk) 08:15, 16 November 2018 (UTC)
Thank you, I very much suspect the text that I copied is bogus, I will remove my copy and the original now.Spitzak (talk) 19:55, 16 November 2018 (UTC)
This article still needs considerable cleanup, as it continuously talks about *all* base-10 systems rather than the Arabic digits. This is making lots of confusing text, such as the sentence I deleted that seemed to claim that these digits are called "Hindu" by Arabs. Actually a careful reading shows that it claims that Arabs call the entire set of base-10 systems "Hindu" but I suspect that is incorrect and the term "Hindu" explictly means not the Arabic digits.Spitzak (talk) 20:10, 16 November 2018 (UTC)
Had a little trouble as somebody thought I was making unwanted changes, despite the fact that I was reverting something I wrote. This time I did not delete the "Hindu Numerals" mention but instead clarified what the article was claiming which is that it is the (translated?) Arabic name for all base-10 systems, including Western and Eastern Arabic digits and also all the Indian ones. I still feel this is probably in error, from your statement it sounds like this term means *only* the Eastern Arabic Numerals. Can you confirm this or otherwise explain what the term means? I do think it would be nice to fix this error.Spitzak (talk) 23:08, 1 December 2018 (UTC)
Yeah, well, the OP didn't give any sources and it doesn't need to be taken seriously. Scholars state that the "Hindu numerals" were received as a package in the Arabic east as well as the Arabic west, and later they evolved through the centuries via differences in hand-writing. There used to be an old theory that the numerals in the Arabic west had some other origin, and they were supposedly called "ghubar numerals". But modern scholars don't agree with these theories any more.[3]
When you are reverting your own edits, please label them as self-revert or some such thing so that we know. -- Kautilya3 (talk) 02:03, 2 December 2018 (UTC)
Actually the Indian numbers Arabs use look different then the original Indian numbers (of which different forms existed), just like the Western numbers (of which different forms existed) actually look different then the Arab numbers (of which different forms existed). Arabs call the Western numbers just that - Western or European numbers.

And the first description of the Indian Numbers was written by a Syro-Aramaeic monk named Severus Sebokht.-- (talk) 22:18, 5 September 2019 (UTC)


  1. ^
  2. ^
  3. ^ Kunitzsch, Paul (2003), "The Transmission of Hindu-Arabic Numerals Reconsidered", in J. P. Hogendijk; A. I. Sabra (eds.), The Enterprise of Science in Islam: New Perspectives, MIT Press, pp. 3–22, ISBN 978-0-262-19482-2

move to 'digits'?[edit]

Shouldn't this be at 'digits', given that 'numeral' means either a part of speech (e.g., in English, 'twelve' is a numeral) or a written number (e.g., 2019 is a numeral, composed of 4 digits, that represents the number two thousand nineteen). A 'digit', on the other hand, is a 'figure' (i.e., a glyph) that indicates a number. 0–9 are digits is this system. — kwami (talk) 02:16, 6 April 2019 (UTC)

From a purely technical and prescriptive perspective, you have a good point, but "Arabic numerals" is a commonly used phrase in English, and "Arabic digits" is not. The average Wikipedia reader wanting to learn more about Arabic numerals is certainly going to end up here first, not on Hindu–Arabic numeral system. And if you put the emphasis on "digits", it makes it sound like Arabic digits could appear mixed together with Roman digits in a single numeral, for instance. (And note the awkwardness of that use of "Roman digits" rather than the oft-used "Roman numerals".) There's been a redirect to this article from Arabic digits since 2004, so anyone thinking along the same bottom-up lines as you are is covered. So in summary, I would say: no. --Dan Harkless (talk) 22:09, 14 April 2019 (UTC)

File:Apices du moyen-âge.PNG nominated for deletion[edit]

FYI, fromâge.PNG:

Metadata claims the image is from 976, which is clearly bogus, since it contains dates from after that, along with modern typography. It appears to be by the anonymous author of the now-dead website As I pointed out in an edit comment in 2017, the rather prominent article en:Arabic numerals has been using the image for some time with text that misleadingly implies it (like the table above it) was created by Jean-Étienne Montucla in the 1700s. This is a useful image, however, and I'd like to see it changed to a lo-res version with corrected metadata (and an intro on en:Arabic numerals that makes its provenance clear), rather than outright deletion. Dan Harkless (talk) 08:33, 10 April 2019 (UTC)
To be clear: I expressed my opposition to the deletion request, but I also don’t think that this file adds anything worthy to this article. It should be properly vetted (I suspect it might be a 19th cent. hoax of some sort) first, and only then included in the article, but with suitable commentary and articulated with the article’s contents. As it is, simply pasted at the end of a section without any explanation for these outlandish glyphs, it should be simply removed from the article — even if not deleted from Commons. Tuvalkin (talk) 01:45, 17 June 2019 (UTC)
Thank you for throwing in your opinion on this long-neglected issue. I've removed the image from the article. I, too, am an inclusionist, and I agree there's no reason for the image to be removed from Commons, nor converted to a fair-use lo-res version as I requested, if it's really an 1800s hoax rather than an image created and copyrighted by the erstwhile webmaster of Just needs to have its very incorrect metadata fixed, if your guess is correct. --Dan Harkless (talk) 06:00, 17 June 2019 (UTC)

A Commons file used on this page has been nominated for deletion[edit]

The following Wikimedia Commons file used on this page has been nominated for deletion:

Participate in the deletion discussion at the nomination page. —Community Tech bot (talk) 09:51, 10 April 2019 (UTC)

Recent edit warring[edit]

I actually thought the anecdote about rules being named wrong was interesting, but this whole thing is in need of some serious cleanup.

This article is about the digits that look like this: 0123456789. IT IS NOT ABOUT BASE-10!!!! Huge amounts of this need to be deleted. Certainly the digits were designed to represent numbers in base-10, so there should be some mention. But there appear to be some editors who are insulted if the fact that base-10 was developed in India is not mentioned as many times as physically possible, thus interrupting the story with developments that (I think, it is really unclear) were done with other forms of the digits.

This has muddled this article considerably by confusing the development and distribute of the idea of base-10 notation with the development of these digits. And even careful reading still seems to show claims that the digits were developed in North Africa, Persia, Spain, and (at least for zero) India. If there is disagreement about the origins this needs to be stated clearly, with all the origin locations together.

Article needs to clearly state that other digits were used to make base-10, and then stick to the development and distribution of glyphs that evolved into the current numerals. And you can direct people to the hindu-arabic-numeral system if they want the history of base-10.

Spitzak (talk) 18:05, 3 May 2019 (UTC)

I have cleaned up some of the POV-pushing, and expanded the section on the origin of Arabic numeral symbols. -- Kautilya3 (talk) 17:00, 4 May 2019 (UTC)
Looks great! Thank you!!Spitzak (talk) 19:04, 4 May 2019 (UTC)

Can numerals be "Hindu"?[edit]

Most western historians translating the Arabic terms translate them as "Hindu numerals". But the proper translation would be "Indian numerals". Hindī is an adjective derived from Arabic Hind (India).

"Hindu" is an older Persian term, which referred to the people of India. It would be odd indeed to call numerals "Hindu".

So, if there are no objections, I would like to change all the occurrences of "Hindu" to "Indian", at least when translating Arabic terms and phrases. (But I would leave alone the "Hindu-Arabic numerlas" term, even though it suffers from the same problem.) -- Kautilya3 (talk) 17:22, 4 May 2019 (UTC)

We need to follow what the sources say and do, whether we agree with them, or not. Paul August 19:29, 4 May 2019 (UTC)
You can find sources for both. We need to make a decision. -- Kautilya3 (talk) 20:28, 4 May 2019 (UTC)
If, as you say, most historians use the term "Hindu numerals" then I would think we should follow that. Paul August 20:53, 4 May 2019 (UTC)
Note that I said I don't propose to change the official name "Hindu-Arabic numerals". But in the English descriptions, we have no reason to follow the ill-informed language of the sources. Here is an Encyclopedia article that does exactly what I recommended:
-- Kautilya3 (talk) 23:17, 4 May 2019 (UTC)
It's not our place to judge sources as "ill-informed". But why don't you list the specific changes you wish to make, and we can discuss them, with sources, individually. Paul August 23:55, 4 May 2019 (UTC)
Charles Burnett is a Professor of the History of Islamic Influences in Europe. I have cited an uptodate WP:TERTIARY source, which can be used to decide DUE WEIGHT when there are disagreements among sources. I will follow this terminology. -- Kautilya3 (talk) 08:57, 5 May 2019 (UTC)