Look closely at the bottom of the Romanian letters ș and ț. The little mark hanging below is a comma, attached to the letter by a tiny gap. That is the correct, standard Romanian form. What you will see almost everywhere — on billboards, in PDFs, in software menus, in text messages — is instead a cedilla: ş and ţ, the hook-shaped mark you know from French ç. The cedilla forms are technically wrong. They are not Romanian letters at all; they are letters borrowed from Turkish that infiltrated Romanian text through decades of bad fonts, limited keyboards, and broken character encodings. This page is your guide to telling them apart and typing the right ones — and it is the one place in this guide where we deliberately display the wrong cedilla forms, clearly labeled, so you can recognize them in the wild.
Comma vs cedilla: the visual difference
A comma below floats just under the letter with a small gap, like a punctuation comma flipped down. A cedilla is a hook that curls and attaches directly to the bottom of the letter (the same mark as in French garçon). On many screen fonts the difference is subtle, but it is real, and at the data level — the actual bytes in a file — they are completely different characters.
| Correct (Romanian) | Wrong (Turkish cedilla) | |
|---|---|---|
| "sh" letter | ș / Ș (comma) | ş / Ş (cedilla) |
| "ts" letter | ț / Ț (comma) | ţ / Ţ (cedilla) |
| Unicode (lowercase) | U+0219, U+021B | U+015F, U+0163 |
| Unicode name | LATIN ... WITH COMMA BELOW | LATIN ... WITH CEDILLA |
| Belongs to | Romanian | Turkish |
So the word for "good morning," correctly written, is bună dimineața — and the cedilla version bună dimineaţa (with ţ) is wrong, even though it looks almost the same and your phone might autocorrect to it.
✅ Și-a pus cizmele și a ieșit în zăpadă.
He put on his boots and went out into the snow. (correct comma-below ș in Și, cizmele, ieșit)
❌ Şi-a pus cizmele şi a ieşit în zăpadă.
Wrong glyphs — these are Turkish cedilla ş; the Romanian letter is ș with a comma below.
How the cedilla crept in
The cedilla problem is an accident of computing history, not a Romanian decision. When character encoding standards were first written in the 1980s and 1990s, Romanian's comma-below letters were not given their own code points. The early Latin-2 and even early Unicode tables only had the cedilla versions (which Turkish needed), so Romanian was told to "borrow" ş and ţ as a stopgap. Fonts, keyboards, and operating systems followed suit. The correct comma-below characters (U+0218–U+021B) were only added to Unicode in 1999 and took years to reach mainstream fonts and keyboard layouts.
The result: an entire generation of Romanian digital text was written with the wrong letters, and the habit stuck. Even today you'll find the cedilla forms in older documents, on signage made with legacy software, and whenever someone types on a layout that hasn't been updated. It is the single most common orthographic error in Romanian — committed not by learners, but by the technology itself.
Multe documente vechi folosesc din greșeală cedila turcească.
Many old documents mistakenly use the Turkish cedilla. (correct ș in greșeală, ț not needed here)
Why it actually matters
It is tempting to shrug — they look almost the same, surely "close enough"? It is not close enough, for three concrete reasons.
1. Search and matching break. To a computer, ş (U+015F) and ș (U+0219) are as different as a and b. Search for "București" (correct) in a document that wrote "Bucureşti" (cedilla) and you may get zero results. Databases, find-in-page, and autocomplete all treat them as unrelated characters.
2. Sorting breaks. Mixed comma/cedilla text sorts inconsistently, because the two characters live in different Unicode ranges. A name list can scatter the same surname into two places.
3. It signals carelessness. In professional, official, and published Romanian, the comma-below forms are the standard, and using the cedilla marks a document as old, machine-mangled, or sloppily produced. For anything that matters — a CV, a website, a printed sign — the cedilla is a visible error.
Caut «Iași» dar fișierul scrie «Iaşi» cu cedilă, deci nu găsesc nimic.
I search for 'Iași' but the file writes 'Iaşi' with a cedilla, so I find nothing. (correct ș in Iași, fișierul)
How to type the correct letters
Getting the comma-below forms is mostly a matter of using the right keyboard layout.
- Windows — Install the "Romanian (Standard)" layout (not "Romanian (Legacy)," which produces the old cedilla forms). On the Standard layout, dedicated keys give you ș, ț, ă, â, î directly.
- macOS — Add the Romanian – Standard input source. The comma-below letters are on dedicated keys; on the U.S. layout you can also use the AltGr/Option-based combinations or character viewer.
- Linux — Choose the "Romanian (standard)" layout in your keyboard settings; AltGr+s and AltGr+t give the comma-below letters on many variants.
- Phones (iOS / Android) — The Romanian keyboard produces the correct comma-below forms. Long-press s and t on a generic Latin keyboard may, on some systems, still offer only the cedilla versions, so a proper Romanian keyboard is safest.
- Unicode entry — As a fallback you can insert U+0219 (ș), U+021B (ț), U+0218 (Ș), U+021A (Ț) directly.
A practical warning: many systems label an older layout as just "Romanian" or "Romanian (Legacy)." That legacy layout produces the cedilla forms. Always pick the one marked Standard.
Pe tastatura «Română (Standard)» obții direct ș și ț corecte.
On the 'Romanian (Standard)' keyboard you get the correct ș and ț directly. (correct comma-below ș, ț)
A quick self-check
If you're not sure which form a document uses, copy a ț or ș and check its Unicode code point (most text editors and online tools will show it). U+0219 / U+021B (and their capitals U+0218 / U+021A) are correct; U+015F / U+0163 are the cedilla intruders. When in doubt, retyping on a Standard layout is the cleanest fix.
Aceasta este forma corectă: ț (virgulă), nu ţ (cedilă).
This is the correct form: ț (comma), not ţ (cedilla). (the second glyph is the Turkish cedilla, shown here only as the wrong form)
Common Mistakes
Using the Turkish cedilla because the keyboard or font defaults to it:
❌ Bucureşti, Iaşi, piaţă, dimineaţa
Incorrect — these use the cedilla ş/ţ; the correct letters are comma-below.
✅ București, Iași, piață, dimineața
Bucharest, Iași, market/square, morning (comma-below ș, ț)
Treating comma and cedilla as "the same letter, just a font choice":
❌ assuming ş and ș are interchangeable
Incorrect — they are different Unicode characters (U+015F vs U+0219); search and sorting treat them as unrelated.
✅ ș = U+0219, the only correct Romanian form
the 'sh' letter, comma below
Selecting the "Romanian (Legacy)" keyboard layout:
❌ typing on 'Romanian (Legacy)' and getting ţ, ş
Incorrect layout — Legacy produces cedilla forms; switch to 'Romanian (Standard)'.
✅ 'Romanian (Standard)' → ț, ș
the layout that produces the correct comma-below letters
Dropping the diacritic entirely to avoid the comma/cedilla question:
❌ Bucuresti, Iasi, piata
Incorrect — plain s/t loses the sound and is only acceptable in casual chat, never in writing that matters.
✅ București, Iași, piață
Bucharest, Iași, market — see 'Why diacritics matter'.
Key Takeaways
- The correct Romanian letters are ș and ț with a comma below (U+0219, U+021B; capitals U+0218, U+021A).
- The lookalikes ş and ţ are Turkish cedilla letters (U+015F, U+0163) — wrong for Romanian, no matter how common.
- The cedilla crept in from legacy encodings, fonts, and keyboards; the correct comma forms were only added to Unicode in 1999.
- "Close enough" fails: they are different characters, so the cedilla breaks search, sorting, and matching and looks unprofessional.
- Type them correctly by using the "Romanian (Standard)" keyboard layout — avoid anything labeled "Legacy."
Now practice Romanian
Reading grammar gets you part of the way. The exercises are where it sticks — free, no signup needed.
Start learning Romanian→Related Topics
- The Romanian AlphabetA1 — Romanian uses the Latin alphabet plus exactly five diacritic letters — ă, â, î, ș, ț — that are full letters in their own right, not optional accents. The result is a 31-letter alphabet in which k, q, w, and y appear only in loanwords, and in which ch and gh are pure spelling devices for keeping c and g hard before e and i.
- Why Diacritics Matter in RomanianA1 — Romanian diacritics are obligatory, not decorative. Dropping them doesn't just look careless — it changes words: peste (over) vs pește (fish), fata (the girl) vs față (face), tata (dad) vs tată (father), mana (manna) vs mână (hand). Diacritic-free Romanian is ambiguous, decodable only from context, and acceptable in casual texting but never in writing that matters.
- The î vs â Spelling RuleA1 — The letters î and â spell exactly the same sound, /ɨ/ — so you never choose between them by ear. The rule is purely positional: write î at the start of a word, at the end, and right after a prefix (în, începe, întâi, neînțeles, a coborî); write â everywhere inside a word (când, mâine, român, pâine). This convention was abolished in 1953 and reinstated in 1993, which is why older texts spell everything with î.
- Consonants: OverviewA1 — Most Romanian consonants have familiar English or Romance values; the signature special letters are ș /ʃ/ and ț /ts/, the always-pronounced h, the j /ʒ/ of 'measure', and a tapped/trilled r that is never the English approximant — getting r and h right strips away most of a foreign accent.
- Romanian Pronunciation: OverviewA1 — Romanian spelling is highly phonemic — you read what you see — so pronunciation is mostly a matter of learning a handful of special letters: the five diacritics (ă, â, î, ș, ț), the soft/hard rule for c and g, and the two central vowels (ă, î/â) that English lacks. This page is the map: the seven vowels, the special consonants, the diphthongs ea/oa, palatalization, and where the stress falls, with a preview of the sounds English speakers find hard.