Why doesn’t fancy text work in Russian, Arabic or Chinese on iPhone?
Bold, script and the rest of iPhone’s fancy text styles are separate Unicode characters standing in for the plain Latin alphabet, and that alphabet is the only one most of them were ever given. Type in Cyrillic, Arabic, Hebrew, Devanagari, Thai or Chinese and there is nothing to substitute at all — not a partial gap like an accented é, but the whole word passing through untouched.
Style an English word in bold or script on a fancy text keyboard and it transforms letter for letter. Try the same thing on привет, مرحبا, שלום, नमस्ते, สวัสดี or 你好 and nothing happens at all — not one character changes, no matter which style you pick or which app you try it in. It is not a bug that only shows up in some languages. It is the plainest version of a gap that this site has already covered for accented Latin letters like é, except here the gap swallows the entire word instead of one letter in the middle of it.
The alphabet behind fancy text was never bigger than Latin
Bold, italic, script, double-struck, circled and the rest of iPhone’s fancy text styles are not fonts a keyboard switches between. Each one is a separate, fixed set of Unicode characters that happens to already look bold or italic or circled, sitting in its own dedicated block of the standard. When that block was created, it was given a-z, A-Z and, for some styles, 0-9 — the plain Latin alphabet a page of mathematics is usually written in, since these blocks exist mainly so a bold vector and an italic scalar can be told apart in an equation, not so a Instagram bio can look different. Nobody scoped it against Cyrillic, Arabic, Hebrew, Devanagari, Thai or Chinese, because none of those scripts were the ones the block was built to serve.
Two different reasons, depending on the script
Cyrillic, Arabic, Hebrew, Devanagari and Thai are alphabets or abugidas of a similar size to Latin — a few dozen letters, not thousands. There is no structural reason a bold Cyrillic block could not exist; it simply never got assigned one, the same kind of gap as the accented Latin letters, just applied to an entire script instead of a handful of diacritics.
Chinese, and the kanji inside Japanese, are a different problem entirely. Han characters number in the tens of thousands, with new combinations still in active use. Giving every one of them a bold duplicate, and then an italic duplicate, and a circled duplicate, would multiply Unicode’s size by however many fancy styles exist, for a font-level distinction that a real bold or italic typeface already handles for free. It is not an oversight waiting to be fixed. Duplicating an alphabet of 26 letters twelve times is a rounding error; duplicating tens of thousands of Han characters twelve times is not.
What actually happens when you try it
A fancy text generator, on iPhone or anywhere else, works the same simple way regardless of which language you feed it: look at each character, check whether the current style has a matching substitute character assigned to it, and swap it in if so. For a-z and A-Z, nearly every style has an answer. For a Cyrillic, Arabic, Hebrew, Devanagari, Thai or Han character, none of the mainstream fancy styles have anything to substitute, so the honest and near-universal outcome is that the original character passes straight through unchanged. The word is not corrupted, deleted or replaced with a box — it is simply left exactly as you typed it, sitting there in its ordinary form while any Latin letters mixed into the same sentence style around it.
- A sentence that mixes scripts styles unevenly. “你好, 𝓭𝓾𝓭𝓮” bolds the English half and leaves the Chinese half untouched, in the same message, because the two halves are being run through completely different parts of Unicode.
- Punctuation and numbers follow their own rule, separate from the script question — the digits 0-9 have matching characters in several fancy styles regardless of what language surrounds them, while ordinary punctuation like commas and question marks is typically left alone everywhere.
- A web generator that claims to offer “fancy Arabic” or “fancy Cyrillic” fonts is usually doing something else entirely — borrowing lookalike symbols from unrelated Unicode blocks rather than a genuine bold or italic version of the alphabet, which tends to render worse and less predictably than the Latin styles it is copying the idea from.
Why this is not the same gap as the accented-letter one
The separate guide on this site about accented letters like café covers a narrower version of this same root cause: Latin fancy text blocks cover the plain 26 letters but never got accented variants, so one letter out of a mostly-styled word stays plain. Cyrillic, Arabic, Hebrew, Devanagari, Thai and Chinese hit a bigger wall for a simpler reason — they were never Latin to begin with, so there was no partial coverage to fall back on in the first place. A French word loses its é; a Russian word loses all of itself.
Which of the 30 typing languages this actually touches
Of the languages a keyboard can typically add on iPhone, this affects Russian, Ukrainian and Serbian written in Cyrillic; Arabic and Persian; Hebrew; Hindi; and Thai, Korean, Japanese and Chinese — each written in a script fancy text styling has nothing built for. English, Spanish, Portuguese, French, German, Italian, Dutch, Polish, Turkish, the Nordic languages, Vietnamese, Greek, Romanian, Indonesian and Serbian written in Latin all sit on the Latin side of the line, where fancy styling works the normal way, accented-letter gaps aside.
Where Kibo fits
Kibo carries all 12 fancy text styles on the keyboard itself, generated as you type, and it also carries Russian, Arabic, Hebrew, Hindi, Thai, Korean, Japanese and Chinese among its 30 typing languages — two separate features that simply do not overlap the way styling and Latin text do. Switching a fancy style on while typing in one of those languages does exactly what is described above: the script passes through in its normal, correct form rather than being mangled, dropped, or swapped for a box, the same principle behind Kibo leaving an unconvertible accented letter alone instead of stripping its accent. None of it needs Full Access, which Kibo reads only for the Copy word, Copy line and Paste buttons on its own toolbar. It is free to download.
Frequently asked questions
Why doesn’t bold or fancy text work when I type in Russian, Arabic or Chinese?
Fancy text styles are separate Unicode characters that were only ever assigned for the plain Latin alphabet, mainly because these Unicode blocks exist for mathematical notation. Cyrillic, Arabic, Hebrew, Devanagari, Thai and Chinese characters have no matching bold, italic or other styled version anywhere in Unicode, so a fancy text tool has nothing to substitute and leaves the original character exactly as typed.
Is this the same problem as accented letters not turning bold?
Same root cause, bigger scale. An accented Latin letter like é is missing from a block that otherwise covers the rest of that word. A Cyrillic, Arabic or Chinese word has no coverage in that block at all, so the whole word stays plain rather than just one letter in it.
Will Chinese characters ever get their own bold Unicode block?
Very unlikely. Han characters number in the tens of thousands, and duplicating that many characters for every fancy style would balloon Unicode’s size for a purely visual distinction that an actual bold typeface already provides for free. It is a scale problem, not an oversight.
Try it on your own keyboard
Kibo is a free custom keyboard for iPhone with real mechanical key sounds, a wardrobe of themes and RGB backlight effects. No account, no ads, and it works with Full Access left off.