EN (English)
Runiform scripts
Rovás
Two scripts and three thousand miles. An angular alphabet engineered in the eighth century to write Turkic vowel harmony on the Mongolian steppe, carried west by migrating confederations, and rebuilt in the Carpathian Basin into a near-perfect fit for Hungarian — then banned, and kept alive on shepherds' tally sticks.
- 47
- letters with pages
- 2
- Unicode blocks
- 8th c.–now
- in use, with a long gap
The alphabet
The Székely-Hungarian Rovás inventory: fourteen vowels, twenty-five consonants, eight syllabic clusters, and the tally numerals. Old Turkic, its ancestor, is covered in the prose below rather than given a row of its own.
Glyphs are shown in Unicode and laid out left to right. Both scripts are written right to left, so these rows are a convenience rather than a facsimile.
Vowels
1–14Fourteen vowel letters in modern use. The historical script did not distinguish long from short, so seven of these are a modern revival convention laid over scribal variants — a point the pages here keep visible rather than smoothing over.
𐳉
3
e
e /ɛ/
The vowel historically omitted more than any other, because every consonant name already contained it.
𐳝
9
ö
ö /ø/
Two scribal variants exist: the Nikolsburg form and the Rudimenta form. Neither is more correct.
𐳭
14
ű
ű /yː/
The awkward one. Unicode has no dedicated long-ü character, so the Rudimenta variant of ü carries it by convention.
Consonants
15–39Twenty-five consonants, one per phoneme. This is the script's whole argument: where Latin needs cs, gy, ly, ny, sz, ty and zs, Rovás has a single character each, and two k's left over from Turkic vowel harmony.
𐳎
21
egy
gy /ɟ/
A voiced palatal stop, one letter, where Latin needs a digraph that suggests the wrong sound entirely.
𐳓
24
ek
k /k/
The front-vowel k. Its twin ak is the last living trace of Old Turkic vowel harmony in this script.
𐳔
25
ak
k /k/
The back-vowel k, and the single most historically interesting letter in the alphabet.
𐳗
27
ely
ly /j/
A letter for a sound Hungarian no longer distinguishes: ly and j have merged in speech.
𐳢
32
er
r /r/
A short-r variant exists at the next codepoint, a scribal rather than phonemic difference.
𐳤
33
es
s /ʃ/
Hungarian s is the sh-sound, which trips up every reader coming from Latin orthography.
𐳥
34
esz
sz /s/
The plain s-sound, written with a digraph in Latin and one letter here. The reverse of what English expects.
Capita dictionum
40–47Syllabic clusters recorded in late medieval alphabet lists, compressing several sounds into one stroke. Their linguistic origin is genuinely obscure and the pages say so rather than inventing one.
The tally numerals
48–53Not letters. An additive number system for counting livestock and debt on split wooden tally sticks, shaped after fingers and crossed thumbs. Explained on this page rather than given pages of their own.
𐳺
48
egy
1 —
𐳻
49
öt
5 —
𐳼
50
tíz
10 —
𐳽
51
ötven
50 —
𐳾
52
száz
100 —
𐳿
53
ezer
1000 —
Where we know it from
Monumental steles, gold vessels, a divination manuscript, an incunabulum endpaper, and a church ceiling.
Kül Tigin and Bilge Khagan steles
Orkhon Valley, Mongolia
732 and 735
Monumental memorials to a Göktürk prince and his brother the khagan, bearing Old Turkic on some faces and Middle Chinese on another. The bilingualism is what made decipherment possible; the content is state propaganda warning against Chinese assimilation.
Tonyukuk inscription
Mongolia
c. 716
Autobiographical military history by the strategist who served three khagans: forced marches through snow in the Altun mountains, night raids, and intelligence from spies. It is also the one Old Turkic inscription that reads left to right.
Irk Bitig
Dunhuang, Gansu
9th century
The Book of Omens: a complete divination manual on paper, in Old Turkic runiform script, from the Uyghur period. A rare case of a runiform script genuinely used for divination — though by drawing lots and reading numbered omens, not by casting letters.
Nagyszentmiklós treasure
Banat
8th–10th century
Twenty-three gold vessels carrying twelve runiform inscriptions, from the Carpathian Basin tradition. Among the richest finds bearing the western branch of the script, and still argued over as to language.
Szarvas bone needle case
Hungary
8th century
An Avar-period needle case with one of the longest Carpathian Basin Rovás inscriptions. An everyday object, which is what makes it valuable.
Nikolsburg Alphabet
Moravia
c. 1483
A complete alphabet list written on the endpaper of an incunabulum: thirty-five letters and fifteen ligatures, with Latin transcriptions. The earliest full inventory of the Székely-Hungarian script, and the source of the Nikolsburg letter variants.
Telegdi, Rudimenta
Hungary
1598
János Telegdi's primer, which codified the alphabet and demonstrated it by transliterating Christian texts including the Lord's Prayer. The source of the Rudimenta letter variants that sit beside the Nikolsburg ones in Unicode.
Énlaka church inscription
Transylvania
1668
A carefully carved Rovás inscription in the coffered ceiling of a Unitarian church. Evidence that the script survived not as a secret but as a craft tradition, applied to a building people worshipped in.
Marsigli's stick calendar
Transylvania
copied 1690
Luigi Ferdinando Marsigli copied out a Székely wooden calendar stick carved in Rovás. The original is lost; the copy preserves a rare example of the script doing the job it was designed for.
Homokmégy quiver fragment
Hungary
10th century
An inscribed bone plate from a quiver, from the period of the Hungarian conquest. Short, contested, and among the earliest candidates for Hungarian written in the script.
Vilhelm Thomsen's decipherment
Copenhagen
1893
The Danish philologist who cracked Old Turkic, working in parallel with Vasily Radlov, using the Chinese faces of the Orkhon steles as a control. He isolated Täñri, Türk and Kül Tigin, then worked out that the script encodes vowel harmony in its consonants.
Klára Sándor
Hungary
modern scholarship
The sociolinguist whose work on the script's history and its modern revival established that today's Rovás is an expanded reconstruction rather than a continuation — new glyphs and rules were added to make an old script fit modern orthography exactly.
Two scripts, one family
This page covers a family rather than a single alphabet.
The Old Turkic script — also called Orkhon, Yenisei or Göktürk — is the earliest writing used for any Turkic language, cut into monumental stone in Mongolia and Siberia from the eighth century. It has around seventy-three encoded characters across its regional variants.
The Székely-Hungarian Rovás is its terminal western descendant, used in the Carpathian Basin and preserved in Transylvania. It is what the word rovás actually names, and it is the alphabet listed in the row above.
Between them lie several centuries and several thousand miles of steppe, and a set of intermediate scripts — Khazarian Rovás, Carpathian Basin Rovás — carried west by the Avars, Khazars and Ogurs.
Not runes
They are called runes constantly and they are not related to the Germanic ones. No shared letter, no shared ancestor, no contact.
The resemblance is a consequence of tools. Both traditions are angular because both were cut into wood, bone and stone with a knife, and a blade makes straight strokes easily and curves badly. Two unrelated writing systems, given the same materials, arrived at the same look.
Where Old Turkic actually comes from is argued. The mainstream account, going back to Thomsen, derives it from Aramaic by way of Iranian intermediaries — most likely Sogdian — carried along the Silk Road; the seventh-century Chinese Book of Zhou notes that the Turks wrote like the Sogdians, which supports the route. A competing tamga theory derives the letters from tribal brands and property marks, pointing out that the sign for oq, arrow, looks like an arrow and the sign for eb, dwelling, looks like a tent. Mainstream linguistics treats the tamga account as unprovable and grants only that the visual conventions of brand-marks shaped how the letters were drawn.
The trick Old Turkic pulls
The Old Turkic script does something no European alphabet does, and it is worth understanding even though the Hungarian descendant abandoned it.
Turkic languages have vowel harmony: a word's vowels are all front or all back, and suffixes shift to match. The script encodes that in the consonants. Most consonants come in two forms, one used only in back-vowel words and one only in front-vowel words. Writing the back-vowel b told the reader that the unwritten vowel beside it had to be a or u; writing the front-vowel b told them it had to be e or ü.
This is why the alphabet has so many letters, and it is why vowels could often be left out entirely: the consonant had already narrowed the possibilities. Thomsen worked this out in 1893 and it is the key that opened the script.
The practical effect is an alphabet shaped precisely around one language family's phonology, and consequently unusable for anything else without modification.
Going west, and changing job
When the script reached the Carpathian Basin with migrating confederations, it met a language it did not fit. Hungarian is Uralic, not Turkic, and handles vowel harmony differently.
So the synharmonic system was dropped — with one survival. The alphabet still has two k letters, ek and ak, front and back, which is the last living trace of Turkic vowel harmony in a Hungarian script. Nothing else remains of it.
What replaced it is a different kind of precision. Hungarian has a large phoneme inventory and Latin serves it badly, needing digraphs — cs, gy, ly, ny, sz, ty, zs — and stacked diacritics. Rovás has one character per sound. Every distinct Hungarian phoneme has its own letter, and the alphabet in the row above is the result.
Vowels you do not write
Historical Rovás was carved onto sticks, and carving takes time, so scribes left vowels out.
The rule was practical. Letters were recited with a leading e — eb, ec, ed, eg, not bé, cé, dé, gé — so the sound e was already implied by naming the consonant, and e and é were dropped almost universally. Other vowels were written only when two fell together, or when one ended a word, or when leaving it out would make the word ambiguous.
Carvers also merged letters. Összerovás, ligaturing, joins the vertical stems of adjacent characters into a single compound sign, compressing a syllable or a short word into one dense shape. Between omission and ligature, a line of historical Rovás is considerably shorter than the same line in Latin, which was the point.
Counting on a split stick
The row above ends with six numerals, and they are not letters.
Rovás numbers are additive, like Roman numerals: larger values first, then smaller, summed. One to four are vertical strokes. Five is a slanted line, read as a thumb held away from the fingers. Ten is a cross, two thumbs or two hands. So 145 is a hundred, four tens and a five.
What makes the system interesting is the medium. Numbers were cut across a stick that was then split lengthwise. Creditor and debtor each kept half, and because the notches ran across the break, neither could add to their own half without the two ceasing to match. It is a tamper-evident contract made of wood, and it is where the word rovás comes from: from róni, to score or notch.
Banned, and then not quite
When Stephen I founded the Christian Kingdom of Hungary in 1000, the state aligned itself with Latin Europe and the Papacy, and the indigenous script was treated as pagan baggage. Its use was suppressed, and records indicate carved sticks were confiscated and burned.
It did not survive in the centre. It survived at the edge, in Transylvania, among the Székely — a Hungarian group serving as autonomous border guards, far enough from the capital to be left alone. There it stopped being a script of administration and became a craft: shepherds' tallies, carved beams, church ceilings, furniture.
By the mid-nineteenth century, with universal Latin-alphabet schooling, genuine everyday use had effectively ended.
The revival, and its politics
Rovás today is more visible than it has been for centuries. Municipalities across Hungary and Transylvania put up bilingual welcome signs with town names in Latin and in right-to-left Rovás. It is taught in clubs, used in folk art, and set in fonts.
Two things should be said plainly about it.
The first is that the modern script is an expanded reconstruction, not a continuation. Klára Sándor and others have shown that historical Rovás did not distinguish long from short vowels and had no letters for foreign sounds. The revival added both: seven long vowels drawn from scribal variants, and Q, W, X and Y built as ligatures. Modern Rovás is engineered to map one-to-one onto modern Hungarian orthography, which the historical script never did.
The second is that the revival has a political dimension, and it is not hidden. The script is used by folklore and tradition groups of every stripe, and it is also used pointedly by right-wing nationalist currents, for whom an alphabet predating Latin Christianity stands for an authentic Hungarian identity uncorrupted by Western Europe. That association is real and it is contested inside Hungary.
As with the Germanic runes, both facts hold at once: a great many people use the script with no political content whatsoever, and the symbolism is nonetheless in play. Knowing that is part of using it knowingly.
Divination
There is nothing here.
Rovás letters have no meanings. They have names — eb, ec, ed — but those are recitation aids, the consonant with a vowel in front of it, not words. There is no rune poem, no kenning tradition, no ideographic layer, and no equivalent of the Stentoften stone using a letter to stand for a concept.
That is a sharper position than the Germanic runes are in. Runic divination is modern, but at least the runes have names that mean cattle, hail and ice, and poems attached to them. Rovás letters are letters and nothing else.
The one genuine divination text in this whole family is the Irk Bitig, a ninth-century Old Turkic manual from Dunhuang. It works by drawing lots and looking up numbered omens — sixty-five short prose passages about hawks, horses and journeys. It is written in the script; it does not read the script.
Frequently asked questions
Latest Posts
Discover the ancient wisdom of tarot cards and unlock the mysteries of your spiritual journey. Our blog is your gateway to understanding the rich symbolism, meanings, and practical applications of tarot reading.
