Why do the Cyrillic 'Z'(З) and the number '3' seem to be the same glyph? Is there a difference that I'm just not seeing? They look identical to me

  • 4
    A Unicode comparison will show that they aren't the same. Just to put them all side by side: 3 З з Jul 7, 2020 at 15:28
  • 11
    @jastako do the letter O and the digit 0 look identical to you? it should be the same with З vs 3
    – ngn
    Jul 7, 2020 at 18:12
  • Visit the Wikipedia article on the letter/sound Z, and take a look at the (animated) picture to the left of the page.
    – Lucian
    Jul 8, 2020 at 1:09
  • @ngn Nice try. O and 0 are OBVIOUSLY shaped differently. Keep in mind letters sometimes look different in different fonts. I created a gif to show the (lack of) difference. drive.google.com/file/d/10Feh2nOlPQL_ZoKpDmh9QCbmc8kZCWTi/…. You can see the difference is barely noticeable only when the are lined up.
    – jastako
    Jul 9, 2020 at 1:05
  • 6
    @jastako they're "OBVIOUSLY" shaped differently in some fonts. There's a reason fonts targetted towards programmers often take extra steps to distinguish between 0 and o, like including a slash or dot in the numeral. Your GIF is inaccessible.
    – muru
    Jul 9, 2020 at 2:21

4 Answers 4


The two are unrelated.

The letter З developed from the Greek letter zeta (Ζ), through an intermediate form with a tail (Ꙁ). This shape got simplified in handwriting until it became the modern form.

The number 3 developed from a Brahmi glyph with three lines, similar to Chinese 三. In cursive writing, this evolved into a modern 3 so that it could be written in a single stroke.

  • 4
    The Brahmi knew Zeta. Resemblance for phonetic likeness is a possibility. One would want /t/ as a prototype--not Samek, which took the same development towards Xi, nor Sade, which looks like Arabic Numeral three now, or Epsilon from He ... if a straight development can be assumed at all, which is but frequently not the case, so that the visual ambiguity to Taw, Tet, X and Chi needs to be considered. It's a mess, like computer fonts before Unicode mixed with 1337-speak, crappy printer drivers, corruptable disks ... or just take the lack of unicode support today for almost all ancient script.
    – vectory
    Jul 7, 2020 at 6:58
  • 5
    Is that Chinese glyph related to Greek Xi? Just kidding :)
    – Keelan
    Jul 7, 2020 at 7:11
  • 1
    @vectory The Greek glyph for 3 was gamma, not zeta, so if the inventors of Brahmi numerals were using it as a base, they would have gone with that instead. Same for classical Cyrillic numbers (gamma was three, zeta was seven).
    – Draconis
    Jul 7, 2020 at 16:14
  • 2
    @vectory Sorry, I have no idea what you're trying to say here. None of the Ancient Greek, Arabic, or Russian words for "three" have a /z/ in them. Most likely zeta came from zayin, not samekh, and its name originally meant "sword". The numeral glyph originated from three lines, not from ṣade. And counting from zero is a much later invention, with no bearing on any of this.
    – Draconis
    Jul 8, 2020 at 16:44
  • 1
    @vectory Wikipedia can answer most of these questions. Zayin is /z/, a voiced fricative. Ṣade is pronounced /ts/ in modern Hebrew, hence its borrowing into Glagolitic (and then Cyrillic).
    – Draconis
    Jul 8, 2020 at 22:08

Draconis explained the origin, but I'll go into another aspect of your question: For practical purposes, yes, it is the same glyph. Native Cyrillic speakers will frequently write the same shape for both when writing by hand, and while I suppose computer fonts have to have a separate glyph for both, I wouldn't be surprised if many font designers copy-paste their design between the two. Even when the design in a font is different, I believe that the average native Cyrillic-written-language speaker will not be able to recognize the difference between the two glyphs if they are shown out of context.

This lack of distinction is so ingrained, it led to amusing situations like frequently hearing people talk about the "Em-pe-ze file format" in the late nineties (I never found out why it wasn't "Em-er-ze", which would have been a more logical mistake to make).

So in general, the glyphs are the same, or perceived as the same. But there are exceptions in some situations.

  • In some handwritings, especially Russian cursive writing, the lowercase letter will have a long tail, like a Latin handwritten "y". Nobody would write the number that way.

enter image description here

  • "Old style" typefaces in Western traditions have numbers which are not aligned with the text glyphs. I am not sure if this was historically used in typography in the Cyrillic writing countries, but nowadays, there are fonts of this style which have Cyrillic letters. Then the number is dislpayed in this style, but the letter is displayed with the other letters.

enter image description here

  • If somebody creates a typeface that tries to emulate the archaic look of the Cyrillic alphabet, both the letter and the number start looking quite different. How different they are are up to the designer; in EvangelieUCS, the numbers become practically unrecognizable, while in Monah OCS they look pretty normal.

comparison two archaic-look fonts

  • 5
    For what it’s worth, I was taught to write a cursive “z” in the same manner you have there—in English. Always confused me, because I don’t think it looks particularly like “z.”
    – KRyan
    Jul 7, 2020 at 13:56
  • I am not sure if this was historically used in typography in the Cyrillic writing countries — I have a russian book printed in 1950, and it uses proportional old style numbers. So, at least one typography was using it.
    – user28434
    Jul 8, 2020 at 8:59
  • @KRyan try imagining a Latin capital Z, or a greek zeta, and then imagine a tiny tail on the lower right corner, as if somebody fused it with a comma. This is something I have seen in handwritten zetas and in archaic-style cyrillic fonts. I don't know if that's how the glyph actually evolved, but to my mind it is a useful association help to understand why the tailed form is related to the capital Z in Latin alphabets.
    – rumtscho
    Jul 9, 2020 at 10:51
  • Thinking about em-pe-ze, shouldn't that have been te-er-ze based on lowercase letter shapes?
    – Jan
    Feb 10, 2022 at 18:01

BTW, this similarity was confusing people in XX century, so they agreed to change the number 3 glyph to slightly differ from the letter. This difference is still maintined in the technical documentation in Russia, as reflected in the GOST standard 2.304-81 note the last glyph of the first two lines - it specifically highlights the number 3:

[GOST 2.304-812

  • 2
    Note that both of the 3-like glyphs in the pictures here represent the digit 3. The letter З differs from the similar version of digit 3 by curving of top and bottom ends of the line towards the middle cusp, and this feature is consistent across all the fonts: types A,B, both slanted and upright.
    – Ruslan
    Jul 7, 2020 at 19:37
  • 2
    My Russian-immigrant high school math teacher (in an honors-level course, Linear Algebra & Analytic Geometry) used to cross her z's, even when printing (not cursive) in the Roman alphabet. A Z looks nothing like a 3 but I guess the habit was too ingrained ... or maybe it was to distinguish Z from 2, just like crossing a 7 is to distinguish 7 from 1? Jul 9, 2020 at 11:30
  • 1
    @RossPresser I do this too. It's to tell Z from 2, which in sloppy handwriting look the same. I haven't realised it wasn't universal, and cursory googling suggests this feature is more common in Europe (just like crossing the 7s) and in maths and engineering.
    – Norrius
    Jul 9, 2020 at 19:10
  • Soviet/Russian post codes are also supposed to use a non-round variant of 3 (codified as GOST R 51506-99), presumably for the same reason.
    – Trang Oul
    Jun 12, 2023 at 11:00

Beside the valid answers given earlier, one more practical point to consider is that mechanical typewriters in USSR (at least the one we had) did not have two separate keys for these, so save some space on mechanics and complexity I guess. So when you needed a "three", you typed "capital Z" and that was it. Surely, in that case the glyphs were identical ;) Elsewhere, a zero and capital O could suffer similarly.

There was also a practice of typing roman numbers with Cyrillic letters, which looks strange to uninitiated, e.g.:

  • I - 1
  • II - П
  • III - Ш
  • V - У
  • X - Х

which led to a number of misinterpretations sometimes.

Rare characters like "L" (roman 50) could be retouched from "1" with a pencil after typing.

More examples of that can be seen e.g. in discussion at https://www.multitran.com/c/m.exe?a=4&MessNum=282414&l1=1&l2=2

  • 6
    For what it's worth, many mid-20th century English typewriters had no number 1 (the numbers would start at two). Typists just typed a lower-case "L". I can remember starting up systems in the 80s and I'd get complaints from users that our systems wouldn't accept numbers with "1" digits in them. I was astounded standing over these folks shoulders, watching them type a mile a minute and realize that they never touched the "1" key.
    – Flydog57
    Jul 8, 2020 at 21:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.