Old Turkic Alphabet (Göktürk Runes)

The grid follows the order used in grammars of Old Turkic: vowels first, then the consonants that go with back vowels (a, ï, o, u), then those that go with front vowels (ä, e, i, ö, ü), then the handful that serve both, and finally the signs that stand for a whole syllable or a consonant cluster. Scholars mark the two consonant series with small numbers: 𐰉 is b¹, 𐰋 is b², 𐰴 is q (the back k) and 𐰚 is k. Each row shows the Orkhon shape from the Mongolian steles and, where the Siberian inscriptions carved a different one, the Yenisei variant next to it.

Only four vowel signs exist, and each covers a pair: 𐰀 is a or ä, 𐰃 is ï or i, 𐰆 is o or u, 𐰇 is ö or ü. The reader knows which because the consonant signs around them already belong to a back or a front series. The script also has single signs for 𐰸 oq/uq, 𐰜 ök/ük, 𐰶 ïq/qï, 𐰱 ič, 𐰰 op/up and 𐱇 ot/ut, for the clusters 𐰡 lt, 𐰦 nt and 𐰨 nč, and the ligature 𐱈 baš 'head', used in a title on the Yenisei stones.

Three signs look like the thing their name means, which is how the first decipherers got a foothold: 𐰸 oq is also the Turkic word for 'arrow' and is drawn as one, 𐰖 ay 'moon' is a crescent, and 𐰋 äb 'house, tent' is a tent with its door. Unicode keeps the scholarly names, so 𐰀 is OLD TURKIC LETTER ORKHON A and 𐰁 the Yenisei form of the same letter.

Who wrote the Göktürk runes: the Orkhon and Yenisei inscriptions

The script belongs to the Göktürks (Kök Türük, the 'Celestial Turks'), whose Second Turkic Khaganate ruled the Mongolian steppe from 682 to 744, and to the Uyghurs and Yenisei Kyrgyz who came after them. The longest and best-known texts stand in the Orkhon valley of central Mongolia: the memorial stele of the prince Kül Tigin (erected 732), that of his brother Bilge Qaghan (735) and that of the old counsellor Tonyukuk (about 725). They are the oldest known texts in any Turkic language, speaking in the first person about the khaganate's wars, its collapse under Tang China and its restoration.

A second family of inscriptions, shorter epitaphs on grave stones along the upper Yenisei in Tuva and Khakassia and in the Talas valley of Kyrgyzstan, was carved in the 8th to 10th centuries by the Kyrgyz and their neighbours. Their letter shapes differ a little, which is why Unicode encodes the Yenisei variants separately. Graffiti and short texts on rock, coins and paper turn up from Mongolia to the Altai, and the script was used in the Uyghur Khaganate until the Sogdian-derived Old Uyghur alphabet replaced it.

Europeans first described the Yenisei stones in the 1720s, and the Russian explorer Nikolai Yadrintsev found the Orkhon monuments in 1889. Nobody could read them until the Danish linguist Vilhelm Thomsen cracked the system in late 1893, starting from the words 𐱅𐰭𐰼𐰃 Tengri 'Heaven' and 𐱅𐰇𐰼𐰜 Türük, and Wilhelm Radloff published the first full translation the next year. Because the strokes look like Germanic runes, the script is often called Turkic runes or runiform, but it has no connection to the Rune Symbols of northern Europe; the likeliest origin is a Sogdian or other Aramaic-derived alphabet reshaped for carving, with a few picture-like signs added.

How the Orkhon script works: direction, vowel harmony and missing vowels

Old Turkic is read from right to left, so 𐱅𐰇𐰼𐰜 begins with the 𐱅 t² at its right end. On the great steles the lines are turned to run vertically, read top to bottom, with columns advancing from right to left; Unicode treats the script as right-to-left, so a line of it pasted into a document will lay itself out that way. Words are separated by a pair of stacked dots like a colon; the ordinary colon or ⁚ (U+205A) does the job in modern text.

The double consonant series exists because Turkic words obey vowel harmony: a word is either all back vowels or all front vowels. So the consonant sign carries that information, and the carvers left out most vowels, writing them mainly at the end of a word or where the reading would otherwise be unclear. 𐰴𐰍𐰣 qağan has no vowel written at all, and 𐰋𐰃𐰠𐰏𐰀 bilgä 'wise' shows its vowels only where they could not be guessed.

To write a modern name, first decide whether its vowels are back (a, ı, o, u) or front (e, i, ö, ü), then take consonants from the matching series and add the vowel signs; today's users usually write every vowel, which the Göktürks would have found wasteful but which makes the result readable. Turkish names and words transfer best, since Turkish still has the same harmony; for other languages choose the series that fits the nearest vowel.

Old Hungarian (Székely rovás): the other runiform alphabet

Old Hungarian, in Hungarian rovásírás or székely-magyar rovás, is the carved alphabet used by the Székelys of eastern Transylvania to write Hungarian from the Middle Ages into the 17th century, and it is listed here because its letters and its right-to-left direction place it in the same runiform family as Old Turkic, probably through steppe scripts that reached the Carpathian Basin with the Magyars. The oldest complete letter list is the Nikolsburg alphabet of about 1483; a church inscription at Csíkszentmiklós is dated 1501; a rovás calendar on a wooden stick was copied in Bologna in 1690; and János Telegdi's Rudimenta of 1598 was a short textbook of the script. Two ö and two ü shapes survive from those sources, which is why the grid has Nikolsburg and Rudimenta forms side by side.

The grid shows the letters in Hungarian alphabetical order, so a reader of the Hungarian Letters & Accents collection will find each sound where they expect it: 𐲀 a, 𐲁 á, 𐲂 b, 𐲄 c, 𐲆 cs, 𐲇 d, 𐲉 e, 𐲋 é and so on to 𐲯 z and 𐲰 zs, with 𐲎 gy, 𐲗 ly, 𐲚 ny and 𐲨 ty as single signs. Old carvings had one case; Unicode gave the script capital and small letters in version 8.0 (2015) because modern writers, signmakers and font designers use both. The numbers 𐳺 1, 𐳻 5, 𐳼 10, 𐳽 50, 𐳾 100 and 𐳿 1000 are tally marks that add up like Roman numerals.

The script is a living revival rather than a museum piece. Since 2010 hundreds of Hungarian towns have put up a second town-name sign in rovás beside the Latin one, it appears on T-shirts, calendars and tattoos, and there are active user communities, keyboards and dozens of fonts. The Unicode name 'Old Hungarian' was itself contested: many users wanted 'Székely-Hungarian Rovás', and you will meet both names, together with 'Hungarian runes', in searches and font names.

How to type Old Turkic and rovás letters: fonts, keyboards and Unicode

Tap any letter above to copy it. Old Turkic has been in Unicode since version 5.2 (2009) in the block U+10C00 to U+10C4F; Old Hungarian followed in Unicode 8.0 (2015) at U+10C80 to U+10CFF. Both are ordinary text and paste into chats, documents and design apps, but neither is part of a default font on most phones and computers, so expect empty boxes until a font is installed. Noto Sans Old Turkic and Noto Sans Old Hungarian from Google are free and complete; on Windows, Segoe UI Historic shows Old Turkic without installing anything. In Microsoft Word, type the code and press Alt+X: 10C00 Alt+X gives 𐰀.

Both scripts are right-to-left in Unicode, so a pasted word reads correctly as long as your editor supports bidirectional text; when mixed into English the Old Turkic word sits inside the line with its own direction. For longer writing, online converters labelled Göktürk, Orkhon or rovás keyboards exist, and the Hungarian rovás community maintains keyboard layouts for Windows, macOS and Android.

Two pitfalls. The Yenisei letters (𐰁 𐰂 𐰄 and the rest) are variants for reproducing the Siberian inscriptions; in a new text, a name or a tattoo, use the Orkhon forms. And because the strokes resemble Germanic futhark letters, images online often mix the two; check that what you copy is from the Old Turkic block and not from the Rune Symbols collection.