R with Accent: ř and ŕ Spell Words, ṛ and ȑ Serve Scholars

Which r with accent belongs to which writing system?

Start from where you met the r with accent. A Czech name, a Sanskrit text and a Serbo-Croatian dictionary each use a different r, and the mark means something different in each. The table sums up Wiktionary's entries for ř (opens in a new tab), ŕ (opens in a new tab), ŗ (opens in a new tab), ṛ (opens in a new tab), ṝ (opens in a new tab), ȑ (opens in a new tab), ȓ (opens in a new tab), ṙ (opens in a new tab) and ṟ (opens in a new tab). The mark names are Unicode's, from its names list (opens in a new tab). Kuvi comes from AccentPad's CLDR facts.

Letter	Mark	Where it is written	What the mark tells you
ř Ř	caron	Czech, Upper Sorbian, Pan-Vlax Romani, Tarifit	a letter of its own, with a different sound in each
ŕ Ŕ	acute	Slovak, Lower Sorbian, Brahui; once Basque	a long syllabic r in Slovak, a soft r in Lower Sorbian
ŗ Ŗ	cedilla, drawn as a comma below	Livonian; formerly Latvian	a soft r
ṛ Ṛ	dot below	IAST and ISO transliteration, Kuvi, Kabyle	the Sanskrit vowel r in IAST
ṝ Ṝ	dot below and macron	IAST transliteration	the long Sanskrit vowel r
ȑ Ȑ	double grave	Serbo-Croatian dictionaries	a short syllabic r with a falling tone
ȓ Ȓ	inverted breve	Serbo-Croatian dictionaries	a long syllabic r with a falling tone
ṙ Ṙ	dot above	ISO 9985, for Armenian	the Armenian letter ռ
ṟ Ṟ	line below	ISO transliteration	ऱ, Tamil ற or Malayalam റ

Read the third column as a warning. Apart from Kuvi and Kabyle ṛ, only the first rows are letters of an everyday alphabet. The others belong to transliteration schemes and dictionaries, so a reader of Czech or Slovak has no reason to know them. The languages card further down reads Unicode CLDR, which lists a main alphabet only for ř, ŕ and ṛ. It leaves the rest out for that reason.

Basque is a footnote. Wiktionary says Sabino Arana's spelling used ŕ for the trilled r, and the standard spelling writes r or rr instead.

Does ř sound the same in Czech and Upper Sorbian?

No. Czech and Upper Sorbian both write ř, the r with caron or hacek, and both spell the word for three tři. They say it differently. These transcriptions are Wiktionary's, from its entries for tři (opens in a new tab), přećel (opens in a new tab) and křidło (opens in a new tab).

Word	Language	Meaning	Said
tři	Czech	three	[ˈtr̝̊ɪ]
tři	Upper Sorbian	three	/ˈtʃi/
přećel	Upper Sorbian	friend	/ˈpʃɛt͡ʃɛl/
křidło	Upper Sorbian	wing	/ˈkʃi.dwɔ/

In each Upper Sorbian word, ř follows t, p or k, and Wiktionary writes it as ʃ, the sh of ship. For the Czech r, Wiktionary writes r̝, a raised trill. The Czech side of the story, its rarity and its history, is on the Czech alphabet page.

Two more languages read the letter in their own way. Tarifit says it as a flap, [ɾ], in the Nador speech Wiktionary cites. Pan-Vlax Romani spelling lists ř as a letter too, with a whole range of sounds. The International Standard for Romani writes rr in its place.

So ř tells you which language you are in before it tells you a sound. The Upper Sorbian name of the letter is eř. Wiktionary gives the Lower Sorbian cousin of křidło as kśidło, with ś where its neighbour writes ř.

Is Slovak ŕ the same letter as Lower Sorbian ŕ?

Same character, U+0155, the r with acute, but not the same letter. In Slovak, ŕ is a long r that carries a syllable; Wiktionary transcribes it /r̩ː/. The Slovak alphabet page explains why a consonant there takes a length mark.

In Lower Sorbian, ŕ is a soft r, /rʲ/, a letter of its own named ejŕ. It does not stay put. Wiktionary's faraŕ (opens in a new tab), priest, a loan from German Pfarrer, shows how: before a vowel the letter becomes the pair rj. So the genitive is fararja. Wiktionary also lists ṙ, r with dot above, as an obsolete Lower Sorbian form of ŕ.

	Slovak ŕ	Lower Sorbian ŕ
Sound	long syllabic r, /r̩ː/	soft r, /rʲ/
Before a vowel	stays ŕ	turns into rj: faraŕ, fararja
Under the language's sort rules	sorted as r	sorted as a letter after r and rj

The last row is our own measurement, not a quoted rule. On 2026-10-10, with Node.js v26.0.0 running ICU 78.3 on CLDR 48 data, Slovak rules matched sŕna with srna when accents were ignored. Lower Sorbian rules kept ŕ apart, and sorted r, rj, ŕ, s in that order.

When does a mark on r make it a vowel?

When the r carries the syllable on its own. The Hrvatska enciklopedija (opens in a new tab) gives the Croatian pairs: r is syllabic in prst, finger, and rt, cape, but not in ruka, hand, or car, emperor. Its neighbours do not fully decide it. In groce, a little throat, r is syllabic, and in grob, grave, it is not.

Many languages have such an r, and they mark it in different ways or not at all. One word shows it. Wiktionary's entries for trn (opens in a new tab) and तृण (opens in a new tab) trace the Slavic word for thorn and the Sanskrit word for grass to one reconstructed root, and list English thorn among the cognates.

Language	Word	Meaning	How the syllabic r is written
Proto-Indo-European	*tr̥nós	(reconstructed)	r̥, r with ring below
Sanskrit, in IAST	tṛ́ṇa	grass	ṛ, the vowel ृ
Slovak	tŕň	thorn	ŕ, because it is long
Serbo-Croatian, in a dictionary	tȓn	thorn	ȓ, for length and falling tone
Czech	trn	thorn	a plain r

In Devanagari the syllabic r is plainly a vowel. Unicode names ऋ DEVANAGARI LETTER VOCALIC R and ॠ VOCALIC RR, and IAST writes them ṛ and ṝ. Wiktionary prints tṛ́ṇa with an acute above the ṛ as well, a second mark on top of the dot.

Czech writes the syllabic r with no mark at all, and so does everyday Croatian. A mark appears where a language has something more to say about that r: its length in Slovak, its tone in a Serbo-Croatian dictionary, or the fact that it is a vowel in a script where vowels are letters.

How do Serbo-Croatian dictionaries mark tone on r?

With the tone marks they put on vowels. The encyclopedia's entry on accent (opens in a new tab) says that standard Croatian combines length and a rising or falling tone. It adds that the tone is usually not written. Phonetic works use four signs: a grave for short rising, an acute for long rising, a double grave for short falling and a circumflex for long falling, as in dànas, lúka, lȕk and lûk.

The syllabic r takes those marks in a dictionary headword. Unicode's names list notes the combining double grave as "Serbian and Croatian phonetics". The headwords below are Wiktionary's; the everyday spelling is the plain word on the left.

Plain spelling	Meaning	Dictionary headword	Mark	Accent
prst	finger	pȑst	double grave	short falling
srce	heart	sȑce	double grave	short falling
vrh	peak	vȑh	double grave	short falling
crn	black	cȓn	inverted breve	long falling
trn	thorn	tȓn	inverted breve	long falling
srna	roe deer	sŕna	acute	long rising

The words come from Wiktionary's entries for prst (opens in a new tab), srce (opens in a new tab), vrh (opens in a new tab), crn (opens in a new tab) and srna (opens in a new tab). Notice the one-syllable words. The encyclopedia says a word of one syllable can take only a falling accent, and pȑst, vȑh, cȓn and tȓn all do.

Two details trip up typists. The encyclopedia draws the long falling tone as a circumflex, while Wiktionary's r headwords use ȓ, an inverted breve, which looks rounder. And the short rising r has no character of its own. In our Python 3.14.6 test, NFC left r plus a combining grave as two code points, and r plus a combining circumflex too. Only ŕ, ȑ and ȓ exist as single characters.

Here the acute means a rising tone. In Slovak it means length. It is the same ŕ, and neither reading carries over.

Which scripts do ṙ, ṛ and ṟ stand in for?

Each stands for one letter of a non-Latin script. Wiktionary names the scheme behind each one, and the Unicode names of the originals come from the names list.

Latin letter	Stands for	Script	Unicode name of the original	Scheme
ṙ	ռ	Armenian	ARMENIAN SMALL LETTER RA	ISO 9985
ṛ	ऋ	Devanagari	DEVANAGARI LETTER VOCALIC R	IAST
ṛ	ड़	Devanagari	DEVANAGARI LETTER DDDHA	ISO
ṝ	ॠ	Devanagari	DEVANAGARI LETTER VOCALIC RR	IAST
ṟ	ऱ	Devanagari	DEVANAGARI LETTER RRA	ISO
ṟ	ற	Tamil	TAMIL LETTER RRA	ISO
ṟ	റ	Malayalam	MALAYALAM LETTER RRA	ISO

Look at the two ṛ rows. In IAST the r with dot below is a vowel, the r of tṛ́ṇa. In ISO transliteration of Devanagari it is ड़, a consonant. A ṛ alone does not tell you which; the scheme the text says it follows does. Wiktionary also lists other uses: a retroflex flap in Americanist phonetics, a half-long r in Inari Sami dictionaries, and a Kabyle consonant.

The vowel has a second spelling. Unicode's names list points readers of ṛ to ISO 15919 on dot below versus ring below, and Wiktionary gives r̥ as a synonym. In our Python 3.14.6 test, ṛ and r̥ stayed different after normalization, so an exact search for ṛṣi does not find r̥ṣi. Under English sort rules in Node.js v26.0.0 with ICU 78.3, at the strength that ignores marks, the two matched each other and plain rsi.

Which old code pages can store ŕ, ŗ and ř?

Only a few, and none stores the scholars' letters. Unicode publishes the byte tables of the legacy code pages. This one comes from the mapping files for Windows-1250 (opens in a new tab), ISO 8859-2 (opens in a new tab), Windows-1257 (opens in a new tab), ISO 8859-13 (opens in a new tab), ISO 8859-4 (opens in a new tab) and Windows-1252 (opens in a new tab).

Code page	ŕ and Ŕ	ř and Ř	ŗ and Ŗ
Windows-1250	E0, C0	F8, D8	none
ISO 8859-2	E0, C0	F8, D8	none
Windows-1257	none	none	BA, AA
ISO 8859-13	none	none	BA, AA
ISO 8859-4	none	none	B3, A3
Windows-1252	none	none	none

None of these files has a byte for ṛ, ṝ, ṙ, ṟ, ȑ or ȓ. Any other r with accent, in a Sanskrit or Armenian transliteration or a dictionary with tones, needs Unicode, which in practice means UTF-8.

The table also explains a familiar kind of damage. A Czech file saved in Windows-1250 and opened as Windows-1252 keeps its bytes but changes its letters, because F8 is ø and E0 is à there.

We decoded a few names that way in Python 3.14.6 on 2026-10-10: Přerov came out as Pøerov, Jiří as Jiøí and Třebíč as Tøebíè. The Lower Sorbian faraŕ became faraà. Reopen such a file as Windows-1250 and the letters come back; retyping them is never needed.