Anatolian Scripts (Luwian Hieroglyphs, Lycian, Lydian, Carian)

Anatolian hieroglyphs are the picture signs carved on rock faces, city gates and stone slabs across Anatolia and northern Syria from about the 14th to the 7th century BC. The oldest are on seals, where they spell a name and a title; the Great Kings of the Hittite Empire put them on monuments such as the rock sanctuary of Yazılıkaya at Hattusa in the 13th century BC, and after the empire fell around 1200 BC the Neo-Hittite kingdoms of Carchemish, Malatya, Karatepe and Hama kept carving them until alphabets pushed them aside in the early 7th century. For a hundred years scholars called them Hittite hieroglyphs. The language they write is in fact Luwian, a close cousin of Hittite spoken across southern and western Anatolia. Hittite itself was written in cuneiform on clay tablets, which you will find in the Cuneiform Symbols collection, and despite the name the hieroglyphs owe nothing to Egypt: they are a native Anatolian invention.

The script mixes two kinds of sign. Logograms stand for whole words and are transliterated with Latin capitals: 𔐑 REX king, 𔖖 DEUS god, 𔔂 URBS city, 𔑺 BOS ox. Syllabograms write one syllable each, a vowel or a consonant plus a vowel, and about sixty of them are enough to spell any Luwian word. Many signs do both jobs: 𔑣 is the foot PES and also the syllable ti, 𔖙 is MAGNUS 'great' and the syllable ur, 𔓊 is LINGUA 'tongue' and la. When one syllable has several signs they are told apart with accents or numbers in transliteration, ta 𔑰, tá 𔐞, tà 𔐬. Monumental texts run in boustrophedon, one line left to right and the next right to left, with the heads of people and animals facing the start of each line, and the signs inside a line are stacked in columns of two to four, with looks sometimes winning over strict reading order.

The script was cracked in stages. A Phoenician–Luwian bilingual found at Karatepe in 1946 by Helmuth Theodor Bossert and Halet Çambel, the inscription of the ruler Azatiwada from the late 8th century BC, gave a long text whose meaning was already known from the Phoenician. In 1973 David Hawkins, Anna Morpurgo Davies and Günter Neumann published the 'new readings', which among other things turned the signs read i and ī (𔖩 𔖪) into zi and za and settled that the language is Luwian. The signs are numbered after Emmanuel Laroche's 1960 sign list, and Unicode kept those numbers: the block Anatolian Hieroglyphs (U+14400 to U+1467F, 583 characters) arrived in Unicode 8.0 in June 2015, and every character is named by its Laroche number, so 𔐀 is simply ANATOLIAN HIEROGLYPH A001.

How to read the famous Luwian signs

A royal inscription opens with 𔐀 EGO, 'I (am)', then the ruler's name spelled out in syllables, then his titles: 𔐑 REX king, 𔐒 MAGNUS.REX great king, 𔐕 HEROS hero, 𔖺 DOMINUS lord, and often 𔓟, the sign for Hatti, the old heartland whose name the Neo-Hittite kings kept using. Gods are written with 𔖖 DEUS in front of them, the way a capital letter marks a name: 𔖖𔓢 is the Storm-god Tarhunza, written with TONITRUS 'thunder' and usually drawn next to 𔓣 his lightning; 𔖖𔓚 the Sun-god and 𔖖𔓜 the Moon-god Arma; 𔖖𔑳 the stag-god Runtiya, whose sign is a stag's head with antlers; 𔖖𔒲 Telipinu and 𔖖𔕥 the Hurrian storm-god Tešub, whose names have signs of their own.

Place names end in 𔔂 URBS for a city or 𔔆 REGIO for a land, so 𔓟𔔆 is 'the land of Hatti'. Genealogies use 𔐰 FILIUS son, 𔕒 NEPOS grandson and 𔕳 AVUS grandfather. 𔔙 DOMUS house, 𔔏 PORTA gate, 𔔭 STELE and 𔔘 AEDIFICARE 'to build' fill the building inscriptions; curses at the end threaten the man who 𔖟 MALUS 'harms' the stone with 𔖖𔓢 and 𔖖𔑳. Numbers are strokes: 𔖭 1, 𔖳 2, 𔖸 3, 𔖼 5, with 𔗁 for 10, 𔗃 for 100 and 𔗄 for 1000, so 𔗁𔖸 is 13 and 𔗃𔗃 is 200. 𔖵 is the word divider some texts use; 𔗍 and 𔗎, the begin and end logogram marks, are editorial marks created for Unicode, never carved on stone.

The syllabary grid at the top gives the sign Luwianists learn first for each syllable, in the order a i u, then ha hi hu and so on down to za zi zu. Several signs serve two vowels at once, 𔖱 ra or ri, 𔗬 wa or wi, 𔗲 la, li or lu, and some sounds have a second sign in common use, 𔐓 á beside 𔗷 a, 𔕐 í beside 𔓯 i. Readings still move: in 2010 Elisabeth Rieken and Ilya Yakubovich showed that the signs long read ta₄ and ta₅ (𔕦 𔓇) are la/i and lá/í, which is why older books spell some names differently from new ones.

Lycian alphabet: 29 letters from the rock tombs of Lycia

The Lycians lived on the mountainous south-western coast of Anatolia, around Xanthos, Patara and Myra, and called themselves 𐊗𐊕𐊐𐊎𐊆𐊍𐊆 Trm̃mili; Hittite texts had known the region as the Lukka lands centuries earlier. Their language, like Luwian a member of the Anatolian branch of Indo-European, was written from about 500 to 200 BC, mostly on the rock-cut house tombs and pillar tombs that still line the cliffs of Lycia. The two long texts are the Xanthos stele, an inscribed pillar with some 250 lines of Lycian, a Greek epigram and a passage in the related Milyan language (Lycian B), and the Letoon trilingual found in 1973 near Xanthos, a decree in Lycian, Greek and Aramaic from 337/336 BC that is now in the Fethiye Museum.

The alphabet grew out of a Greek one with extra letters for Lycian sounds, and runs left to right like Greek. It has 29 letters: the vowels 𐊀 a, 𐊁 e, 𐊆 i, 𐊒 u and the nasal vowels 𐊙 ã and 𐊚 ẽ; the nasal consonants 𐊐 m̃ and 𐊑 ñ beside plain 𐊎 m and 𐊏 n; and a set of consonants whose exact sounds are still debated, written 𐊂 b, 𐊃 β, 𐊄 g, 𐊅 d, 𐊇 w, 𐊈 z, 𐊉 θ, 𐊊 j, 𐊋 k, 𐊌 q, 𐊍 l, 𐊓 p, 𐊔 κ, 𐊕 r, 𐊖 s, 𐊗 t, 𐊘 τ, 𐊛 h and 𐊜 χ. Tomb texts open with a fixed formula, 𐊁𐊂𐊚𐊑𐊏𐊚 𐊜𐊒𐊓𐊙 ebẽñnẽ xupã 'this tomb', followed by who built it and for whom. Inscriptions separate words with a colon of two dots, for which Unicode uses ⁚ (U+205A). The Lycian block (U+10280 to U+1029F) was added in Unicode 5.1 in 2008. Several letters look exactly like Latin or Greek capitals, 𐊁 E, 𐊆 I, 𐊎 M, 𐊏 N, 𐊕 R, 𐊗 T, so never patch a Lycian word with keyboard letters: it will look right and search wrong.

Lydian and Carian: two alphabets that keep surprising scholars

Lydian was the language of the kingdom of Croesus, written in and around its capital Sardis, 𐤮𐤱𐤠𐤭𐤣 Sfard, from about 700 to 200 BC. The earliest pieces are coin legends and graffiti; the longer texts, about 110 in all and most of them tomb inscriptions, come from the 5th and 4th centuries BC under Persian rule, and at least five of them are poetry with a twelve-syllable line and rhyming vowels at the line ends. The script was read thanks to a Lydian–Aramaic bilingual found at the Sardis necropolis in 1912, which Enno Littmann used in 1916 to work out most letters and the outline of the grammar. The alphabet has 26 letters, related to the Greek and Phoenician ones, and is written right to left; Unicode encodes it that way, so a pasted Lydian word such as 𐤤𐤮 𐤥𐤵𐤫𐤠𐤮 eś wãnaś 'this tomb' flows from the right and the cursor jumps as you type. That is the script working, not a bug. Oddities worth knowing: 𐤱 f has the figure-eight shape that also turns up in Etruscan and Oscan alphabets, there are two s letters 𐤮 and 𐤳, two nasal vowels 𐤵 ã and 𐤶 ẽ, and three letters whose sounds are uncertain, 𐤷 λ, 𐤸 ν and 𐤹 c. 𐤿 is a triangular mark that a few texts use like a quotation mark; words are otherwise divided by spaces, which is rare in ancient scripts. The Unicode names keep the older transliteration (b, v, ś, s), while newer grammars write those four letters p, w, s and š.

Carian was spoken in the south-western corner of Anatolia around Halicarnassus, Mylasa and Kaunos, but most of its inscriptions were found in Egypt: about 170 there against some 30 in Caria. Carian mercenaries served the pharaohs from the early 6th century BC, scratched their names on the colossi of Abu Simbel and the fort at Buhen, and were buried at Memphis under stelae in Carian and Egyptian. Those bilingual stelae are what finally cracked the script: in the 1980s John Ray, Diether Schürr and Ignacio Adiego showed that the letters, although they look Greek, mostly do not sound Greek. 𐊠 is a, 𐊫 o and 𐊲 u as expected, but the letter shaped like theta 𐊨 is q, and a whole series of look-alikes have other values. Their system was accepted by 1993 and confirmed in 1996 by a Carian–Greek bilingual from Kaunos. Unicode 5.1 (2008) encodes 49 Carian letters in the block U+102A0 to U+102DF because it unites the regional alphabets of Kaunos, Hyllarima, Mylasa, Memphis and the rest, each of which used around 30; the order follows Adiego's list, so the grid above groups them by sound instead. Two letters, 𐊱 and 𐋆, still have no agreed value and are named only by number. The few words everyone agrees on include 𐊭𐊺𐊢 ted 'father' and 𐊺𐊵 en 'mother'.

How to type Anatolian scripts, fonts and look-alikes

No phone or desktop keyboard layout exists for any of these four scripts, so copying is the practical way: tap a sign above, or on Windows in Microsoft Word type its code and press Alt+X (14400 Alt+X gives 𔐀, 10280 gives 𐊀, 10920 gives 𐤠, 102A0 gives 𐊠). The code is on each sign's own page. The characters are plain Unicode text, so they survive in documents, chats and web pages, but only as long as a font is present.

Fonts are the real hurdle. The free Noto family covers all four: Noto Sans Anatolian Hieroglyphs, Noto Sans Lycian, Noto Sans Lydian and Noto Sans Carian. Windows ships Segoe UI Historic, which draws Lycian, Lydian and Carian but does not list Anatolian Hieroglyphs, so the Luwian signs need Noto there too. Most phones show empty boxes for the hieroglyphs unless the app bundles a font; if a sign shows for you and not for your reader, that is why. For publishing, remember the scholarly conventions: Luwian logograms in capitals (DEUS, REX), syllables in italics with their numbers (tá, ta₄), Laroche numbers with an asterisk (*17 is REX), and for the alphabets the transliterations shown in the grids.

Look-alikes are everywhere here. Lycian 𐊁 𐊆 𐊎 𐊏 𐊕 𐊗 and Carian 𐊠 𐊫 𐊹 𐊺 resemble Latin and Greek capitals but are different characters, and Lydian 𐤠 𐤡 𐤢 share shapes with the Phoenician letters they descend from; keep each word in one script. Related collections: Greek Letters for the parent of all three alphabets, Cuneiform Symbols for the Hittites' own clay-tablet script, Linear B Syllabary & Ideograms for the other Bronze Age syllabary of the Aegean, Old Italic (Etruscan) & Gothic Alphabets for the figure-eight f, and Egyptian Hieroglyphs for the picture script that Anatolian hieroglyphs are so often compared with.