# So here’s a silly question

Hebrew – and Arabic maybe? – read right to left, not left to right.

OK, it’s only a convention after all. And yes, I have noted a bloke reading a Hebrew language book “back to front”, so it’s not just on the one page.

OK, only a convention.

So, what happens with numbers. Is it 001 \$ for a Benjamin? 100 \$? Umm, what? 5141.3 for Pi?

## 33 thoughts on “So here’s a silly question”

1. From what I recall of Jordanian bank notes, they read their numbers little-endian (whereas we read them big-endian)

2. Wiki (I know, not a reliable source), says same order as us, ie highest place- value on the left.

3. Be interesting to know what the historic convention was in Chinese/Japanese, since they write (wrote) vertically. I presume as the characters represent the spoken word, vertically.

4. Wiki is correct about the visual order of ‘Arabic’ numbers, but they’re written (and, presumably read) from R-L. The words for the numbers are similar to one-and-twenty etc, but then hundreds go in front: 221 would be “hundreds two, one and twenty”. The digits 0-9 go: ٠١٢٣٤٥٦٧٨٩ – so 9 is in the right place, but what’s zero doing at 5 and the Arabic zero resembles our decimal point. All a bit weird.

5. I knew Arabic numbers were written the same as ours, i.e. increasing powers of ten to the left. I didn’t know that in speaking them they rendered it right to left. I suppose it’s logical when letter & word ordering is also right to left.

Puzzled about German though. In English it’s twenty two but German is zwei und zwanzig, French is vingt-deux, Spanish is veintidós.

6. When arabic numbers were introduced into Europe we copied them verbatim rather than reversing the digits to allow for the reversed writing that they were embedded in, so in some sense it’s us left-to-right types getting it wrong. Arguably teaching addition to kids would be slightly easier if the carries propagated left to right.

Amusing history: some of the first computers displayed numbers as little-endian wave forms on a CRT so Turing and colleagues got used to adding them up that way:

71+53 = 25

7. I’ve seen written Arabic where the numbers and the writing are opposite, so it’s something like:

750 teg uoy dna 25 dna 30 ylpitlum

Classical Chinese, and Japanese when written in Chinese, is top to bottom, right to left, so:
noitfsR
cnshiee
edterna
.shnstd
.eettet
.nshanh
.teinci
.ecsdes

8. Newsprint and magazines in Japan are still typically top-to-bottom with columns from right to left, although internet editions are horizontal left to right. Company names and slogans on trucks are often nose-to-tail.

9. Funniest thing about Chinese counting is that they put the comma after every four digits not 3. So…basic unit of big numbers is the Wan (10,000). We say one million – they say a hundred Wan. This can get confusing. We say 20 million – they say 2,000 Wan. When you get to Wan x Wan = A hundred million they say Yi. So 245 million for us is 2 Yi 4,500 Wan (si qian wu bai wan). I’ve almost never seen large numbers translated correctly because the fundamental unit and mindset is different – you’re not translating but recalculating.

10. Same in Japan for big numbers but not such a problem in practice since 1oku(億、same as 亿 in commie chinese) ie Y100mn is typically an approximation of USD1mn.

11. Modern Hebrew writes numbers the same way as English.

Classical Hebrew didn’t have numbers.

12. .dnalgnE ni detrats noitasilivic reporp rednow on s’ti ,feirg dooG

13. In India, there’s a comma after three digits from the right, then after every two: 12,34,56,789. Those off us who’ve been to Mumbai can annoy people by pointing out that the numbers in the captions in “Slumdog Millionaire” are wrongly written; also that everyone says “VT”, not “Victoria Terminal”.

14. When I wrote Farsi (which was a very long time ago) I remember writing the digits from the left, though the words went from the right. So I had to know ahead of time how many digits I was writing to know where to start and to make space for it.

15. “In English it’s twenty two but German is zwei und zwanzig”

Consider the lyrics of The Ball at Kirriemuir.

16. As a left-hander I’m always jealous that they get to use a fountain pen without smudging.

17. Classical Hebrew used letters of the alphabet as numbers, so א‎ for 1, ב‎ for 2, and so on. But yeah, Arabic numerals written left-to-right nowadays. They also don’t flip commas around or anything. So yes, it looks a little weird when all of those components are mixed together.

Something I never understood is how vowels are used in Hebrew. We were taught to use vowels when reading religious texts in Sunday school (these are the dots and lines surrounding letters), but if you actually go to Israel or read the Jerusalem Post in Hebrew, no vowels whatsoever. I have yet to figure out how adult Israelis are able to assume pronunciation.

18. I heard a Swiss girl once, on the telephone, speaking German and giving a telephone number.
Say it was 752436.
She said (I translate) “Five and seventy, four and twenty, six and thirty.”
I presume it was taken down correctly.
I also know Hebrew. When I speak English, I visualise the words as written. When I speak Hebrew, I visualise the words written simultaneously in both Hebrew and English letters, going both directions at once and always in the right order.
Odd thing, the brain.

19. I learned Japanese in the 1980s, used it very little for about 30 years. Then on site some years ago I needed to verbally “grab” the team leader before he dashed off into another room, and my mind went blank and all I could think of was the Japanese. Chotto matte… agh, wait! err…

20. English it’s twenty two but German is zwei und zwanzig, French is vingt-deux, Spanish is veintidós.

I mean ok, but the French say four twenties, ten and nine.
So we can’t really take them seriously.

21. I spoke hebrew as a child and Snag is right. Thought that reading right to left and then numbers left to right was stupid. But I am 90% Anglo-Saxon. Sigh.

22. dearieme: I assume that was for the metre rather than standard Scots usage, but then as a Sassenach, what do I know?

23. Fascinating! It’s the gradual shift to positional notation reflected in the language(s). English speakers also talked about four score and seven, six and twenty etc.
One would have thought that right to left writers would do the same with numerical position. What I’m reading is that “western style” dominates but with regional variations as to grouping (how many numerals between commas) persisting.

24. . . . the French say four twenties, ten and nine.

Wot, as in luftballons? Fuck off, launch the missiles.

Four twenties, ten and nine balloons
Floating through the summer sky

Bollocks, that bit really works. Recall the missiles.

And they sort of actually did it, quatre-vingt-dix-neuf ballons:

25. If the duodecimal system was good enough for the Babylonians it’s good enough for me.
02 blackbirds baked in a pie.
FIFY

26. “One would have thought that right to left writers would do the same with numerical position. What I’m reading is that “western style” dominates but with regional variations as to grouping (how many numerals between commas) persisting.”

Or that “westerners”, left-to-righters, have managed to get comfortable with reading numbers the wrong way round. But Israelis are weird.

Reading habits for both words and numbers contribute to the SNARC effect
SAMUEL SHAKI
Ariel University Center of Samaria, Ariel, Israel
MARTIN H. FISCHER
University of Dundee, Dundee, Scotland
AND
WILLIAM M. PETRUSIC

Psychonomic Bulletin & Review
2009, 16 (2), 328-331
doi:10.3758/PBR.16.2.328

This study compared the spatial representation of numbers in three groups of adults: Canadians, who read both English words and Arabic numbers from left to right; Palestinians, who read Arabic words and Arabic–Indic numbers from right to left; and Israelis, who read Hebrew words from right to left but Arabic numbers from left to right. Canadians associated small numbers with left and large numbers with right space (the SNARC effect), Palestinians showed the reverse association, and Israelis had no reliable spatial association for numbers. These results suggest that reading habits for both words and numbers contribute to the spatial representation of numbers.

27. The German word order thing is interesting. Also been research on that, e.g.

https://eprints.whiterose.ac.uk/93082/

Psychol Res 2016 May;80(3):422-33. doi: 10.1007/s00426-015-0729-y. Epub 2015 Dec 15.

Processing multi-digit numbers: a translingual eye-tracking study
Julia Bahnmueller, Stefan Huber, Hans-Christoph Nuerk, Silke M Göbel, Korbinian Moeller

28. Just to add my tuppence worth, I have had Excel spreadsheets sent to me from Arabic sources. The numbers are formatted as we would like to see them but the “start” cell A1 is at the top right hand corner.

29. All these comments, and yet no one has yet pointed out that Pi to 4 decimals is 3.1416 not 3.1415 (as it rounds up from 3.14159) as Tim wrote above.

Dearie dearie me.

30. No reason to. This a discussion about writing numbers. Not mathematics. The fourth decimal place of pi is 5, fifth is 9 but not being written