Home Back In Russian Contents Hu

The Khazar Rovas (KR) script is directed exclusively from right to left; there is no casing and no numerals in any of the known Khazar Rovas relics. In some cases, borrowing characters from other scripts led to the multiplication (duplication) of KR characters (Sec. 3.5.2). Sporadically and not consequently, KR applied vowel harmony in case of some consonants (front consonats used with front vovels, back consonats used with back vovels, i.e ki vs. ɣï (ɣı), qï (qı)).

7.1.2. Khazar Rovas characters

Table 7.1.2-1 details the character repertoire of KR. The number of occurrences of each glyph in each relic is given in brackets at the end of the rows. In the first column the glyphs of the normalized Khazar Rovas characters are listed according to the international standard proposal of the Hungarian Standards Institution.⁴⁶⁸

The notations like the following “in As-Alan, in lig., sound: /a/ (4)” must be viewed critically, for example:
1. As-Alan attribution is frivolous, unjustified, of notional character. Genetic analysis of similar claims demonstrated no connection between imagined “As-Alan” and people buried.
2. The presumption of widespread ligatures in reality stands for “we do not know”
3. Accordingly, the phonetic value may be incorrect
4. The phonetic value must be consistent with the morphology of the script
5. Attribution to “As-Alan” is unprovable because there are no known examples of “As-Alan” language in space and time.

Ditto with attributions Khazar, Kipchak, Ogur, etc. Ditto with dating, none of which is instrumental and all of which are of notional character. In the sources, all attributions and datings are preceded by qualifiers like “probably”, “possibly”, “likely”, and such, and those that are not qualified deserve even less credibility. The proposed reading is erroneous if a scholar needs a ligature for the reading, unless a bi-lingual inscription dictates a suspicion of a ligature.

Table 7.1.2-1: KR characters and examples of each glyph in relics

Glyph	Name	IPA symbol	Examples of the Khazar Rovas glyph in relics
Y	FORKED A	/a/ɛ/	Village Krivy., Sec. 7.2.7, in As-Alan, in lig., sound: /a/ (4) Ermen Tolga, Sec. 7.2.12, in Khazar, sound: /a/ɛ/ (6) Mayak Large, Sec. 7.2.9, in Khazar, in lig., sound: /a/ (3) Mayak Smaller inscriptions, 9th c., Fig. 7.2.10-1,5a, in Khazar, partly in ligature, sound: /a/ (2) Mayak Smaller, 9th c., Fig. 7.2.10-3, in As-Alan, in ligature, sound: /a/ (1)
X	B	/b/	Homokmegy-Halom, 10th c., Sec. 7.2.13, in Turkic (1) Mayak Smaller Inscr., 9th c., Fig. 7.2.10-2, in Ogur (1)
	ANGLED B	/b/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (3) Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (2) Khumara, 9th—10th c., Fig. 7.2.5-3, in Khazar (1) Novocherkassk, 9th—10th c., Sec. 7.2.6, in Kypchak (1) Village Krivyanskoe, 9th— 10th c., Sec. 7.2.7, in As-Alan (1) Ermen Tolga, 8th—10th c., Sec. 7.2.12, in Khazar (8) Mayak Smaller, 9th c„ Fig. 7.2.10-1,5a,b, in Khazar (2) Bilyar, 11th- 13th c., Sec. 7.2.16, in Ogur (l)
	ARCHED B	/b/p/	Jitkov, Bow Cover, 8th c., Sec. 7.2.1, in Khazar, sound: /b/ (1) Mayak Large, 9th c., Sec. 7.2.9, in Khazar, sound: /p/ (1)
	RAISED B	/b/	Khumara, 9th—10th c., Fig. 7.2.5-4, in Ogur (1) Alsoszentmihâlyfalva, 920-952 AD, Sec. 7.2.14, in Khazar (1) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (4) Mayak Smaller, 9th c., Fig. 7.2.10-3, in As-Alan, in lig. (1)
	TRIPLE CS (SH/S)	/tʃ/s/	Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (4) Khumara, 9th—10th c., Fig. 7.2.5-4, in Ogur, in ligature (1) Village Krivyanskoe, 9th -10th c., Sec. 7.2.7, in As-Alan (3) Mayak Large, 9th c., Sec. 7.2.9, glyph: , in Khazar (1)
	ARCHED D	/d/di/dɛ/j/dʒ/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (2) Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (1) Khumara, 9th-10th c., Fig. 7.2.5-lb,2b, in Khazar (3) Khumara, Fig. 7.2.5-4, in Ogur, sound: /di/, in lig. (1) Novocherkassk, 9th-10th c., Sec. 7.2.6, in Kypchak (1) Village Krivyanskoe, Sec. 7.2.7, in As-Alan, in lig. (2) Alsöszentmihâlyfalva, 10th c., Sec. 7.2.14, in Khazar, sound: /ed/, in ligature (1) Ermen Tolga, Sec. 7.2.12, in Khazar, partly in ligature (6) Mayak Large, 9th c., Sec. 7.2.9, in Khazar, partly in lig. (8) Mayak Smaller, Fig. 7.2.10-1, in Khazar, in ligature (1) Mayak Smaller inscr., 9th c., Fig. 7.2.10-3, in As-Alan (1)
	FORKED D	/d/	Homokmegy-Halom, 10th c., Sec. 7.2.13, in Turkic (1)
	SHARP D	/d/ð/dʒ/ (There is a large semantical difference between /d/ and /ð/, in English /ð/ is transmitted as th, which comes in voiced and unvoiced veriety. The Türkic /ð/ is transcribed as d or δ, apparently depicting only a voiced version. In the early Horesmian Türkic alphabet the interdental th /ð/ was depicted as , identical to the Futhark Þþ, which stands for voiced and unvoiced th. This cardinally changes transcription: Ishad becomes Ishath)	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (3)) Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (7) Mayaki Amphora, c., Sec. 7.2.2, in Ogur (1) Devitsa Coins, 8lh-9th c., Sec. 7.2.3, in Khazar, in ligature (2) Khumara, ca. 9th— 10lh c., Fig. 7.2.5-lb,2b,3, in Khazar (5) Khumara, 9th—10lh c., Fig. 7.2.5-4, in Ogur (1) Novocherkassk, 9th—10th c., Sec. 7.2.6, in Kypchak (3) Village Krivyanskoe, 9th—10th c., Sec. 7.2.7, in As-Alan (3) Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic (1) Bilingual Khazar Coin, Sec. 7.2.11, in Khazar, in ligature (1) Ermen Tolga, 8th—10th c., Sec. 7.2.12, in Khazar (1) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (6) Mayak Smaller, 9th c., Fig. 7.2.10-1, in Khazar (1) Mayak Smaller, Fig. 7.2.10-2, sound: /dʒ/, in Ogur (1) Kievan Letter, 955-961, Sec. 7.2.15, in Khazar (1) Bilyar, 11th-13th c.,Sec. 7.2.16, in Ogur (l)
	DIAGONAL E	/ɛ/	Mayaki Amphora, 8th-9th c., Sec. 7.2.2, in Ogur (1) Khumara, 9th -10th c., Fig. 7.2.5-2b,3, in Khazar, glyph: (2) Mayak Large, 9th c., Sec. 7.2.9, in Khazar, glyph: (2)
	ANGLED E	/ɛ/	Achiktash, first half of the 8,h c., Sec. 7.2.4, in Kypchak (2)
	ARCHED E	/a/a:/ɛ/	Jitkov, 8th c., Sec. 7.2.1, in Khazar, glyph: , sound: /a/a:/ɛ/ (5) Khumara, Fig. 7.2.5-1b, 2b, 3, in Khazar, sound: /a/ (5)
	FORKED E	/a/ɛ/	Novocherkassk, Sec. 7.2.6, in Kypchak, sound: /a/ɛ/ (3) Alsöszentmihâlyfalva, Sec. 7.2.14, in Khazar, sound: /a/ (1)
	CLOSE E	/e/	Achiktash, first half of the 8lh c., Sec. 7.2.4, in Kypchak (1)
	CENTRAL F	/f/	Village Krivyanskoe, 9th— 10th c., Sec. 7.2.7, in As-Alan (1)
	CLOSE G	/g/ɣ/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak, sound: /g/ (1) Khumara, 9,h—10th c., Fig. 7.2.5-1b, 2b in Khazar, sound: /g/ (3) Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic, sound: /g/ (3) Homokmegy-Halom, Sec. 7.2.13, in Turkic, sound: /ɣ/ (1) Ermen Tolga, 8th—10th c., Sec. 7.2.12, in Khazar, sound: /g/ (1) Mayak Large, 9th c., Sec. 7.2.9, in Khazar, sound: /g/ (6)
	FORKED G	/g/	Khumara, 9th— 10th c., Fig. 7.2.5-4, in Ogur (1)
N	GH	/ɣ/ in As-Alan languages: /g/	Jitkov, 8th c., Sec. 7.2.1, in Khazar: , sound: /ɣ/ (3) Achiktash, Sec. 7.2.4, in Kypchak, glyph: , sound: /ɣ/ (1) Mayaki, Sec. 7.2.2, in Ogur, calligr., mirrored , sound: /ɣ/ (1) Khumara, 9th— 10th c., Fig. 7.2.5-2b,3, in Khazar, mirrored: , sound: /ɣ/ (3) Khumara, Fig. 7.2.5-4, in Ogur, mirrored: , sound: /ɣ/ (1) Village Krivyanskoe, Sec. 7.2.7, in As-Alan, mirrored: , sound: /g/ (1) Minusinsk, Sec. 7.2.8, in Turkic, mirrored: , sound: /ɣ/ (1) Alsöszentmihâlyfalva, 10th c., Sec. 7.2.14, in Khazar, in ligature, sound: /ɣ/ (1) Mayak Large, 9th c„ Sec. 7.2.9, in Khazar, sound: /ɣ/ (2) Kievan Letter, 955-961, Sec. 7.2.15, in Khazar, sound: /ɣ/ (1)
	TRIPLE GH	/ɣ/	Minusinsk, 8th-9lh c., Sec. 7.2.8, in Turkic, overwritten (2) Homokmegy-Halom, 10th c., Sec. 7.2.13, in Turkic (2)
	ANGLED CH (KH)	/x/	Mayaki Amphora, c., Sec. 7.2.2, in Ogur (1)
	ARCHED CH (KH)	/x/	Village Krivyanskoe, 9th-10th c., Sec. 7.2.7, in As-Alan (1)
	FORKED CH (KH)	/x/	Jitkov, Bow Cover, 8lh c., Sec. 7.2.1, in Khazar, glyph: (1) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (1)
	SHARP CH (KH)	/h/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak, sound: /h/ (4)
	ANGLED I	/i/, in ligature also /e/	Jitkov, Sec. 7.2.1, in Khazar, mirrored glyph (1) Mayaki Amphora, 8lh-9th c., Sec. 7.2.2, in Ogur (1) Khumara, 911—10th c., Fig. 7.2.5-4, in Ogur, in lig., mirrored (1) Mayak Large, Sec. 7.2.9, in Khazar, in lig., sound: /e/ (1) Mayak Smaller, Fig. 7.2.10-4, in As-Alan, in lig. (1) Ermen Tolga, Sec. 7.2.12, in Khazar, in ligature, sound: /e/ (3) Alsöszentmihâlyfalva, 10th c., Sec. 7.2.14, in Khazar, in ligature, sound: /e/ (3) Kievan Letter, Sec. 7.2.15, in Khazar in lig., mirrored (2)
	ARCHED I	/i/ï/	Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (4) Khumara, Fig. 7.2.5-2b, 3, in Khazar, partly in ligature, (2) Village Krivyanskoe, 9lh— 10th c., Sec. 7.2.7, in As-Alan (2) Mayak Large Inscr., 9th c„ Sec. 7.2.9, in Khazar (4) Mayak Smaller, Fig. 7.2.10-2, in Ogur, mirrored: (1) Mayak Smaller. Fig. 7.2.10-5a,b, in Khazar, glyph: (1)
	CIRCLE ENDED I	/i/ï/	Khumara, 9th— 10th c., Fig. 7.2.5-lb, in Khazar, glyph: (1) Khumara, 9th— 10th c., Fig. 7.2.5-4, in Ogur (1) Minusinsk, 8th-9,h c., Sec. 7.2.8, in Turkic, glyph: (2) Ermen Tolga, 8th— 10th c., Sec. 7.2.12, in Khazar (1)
	CLOSE J	/j/	Jitkov, First third of 8th c., Sec. 7.2.1, in Khazar (1) Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (1) Khumara, 9th-10th c., Fig. 7.2.5-2b,3, in Khazar, glyph: 0 (2) Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar (2) Minusinsk, 8lh-9lh c., Sec. 7.2.8, in Turkic (2) Mayak Smaller, 9th c., Fig. 7.2.10-1, in Khazar (1) Mayak Smaller, 9th c., Fig. 7.2.10-4, in As-Alan, in lig. (1)
	FORKED K	/k/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar, in ligature (1) Bilingual Khazar Coin, Sec. 7.2.11, , in Khazar (1) Mayak Smaller, 9th c., Fig. 7.2.10-4, in As-Alan (1) Kievan Letter, 955-961, Sec. 7.2.15, in Khazar (1)
	KUE	/ky/	Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar (1)
	ARCHED L	/l/	Khumara, 9th- 10th c., Fig. 7.2.5-4, in Ogur (1)
	CROSSED L	/l/	Homokmegy-Halom, 10th c., Sec. 7.2.13, in Turkic (1)
	FORKED L	/l/	Minusinsk, 8lh-9th c., Sec. 7.2.8, in Turkic (1)
	SIMPLE L	/l/	Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (1) Mayak Large inscr., 9th c., Sec. 7.2.9, in Khazar (2) Alsöszentmihâlyfalva, Sec. 7.2.14, in Khazar, in lig. (1) Kievan letter, 955-961, Sec. 7.2.15, in Khazar in lig.(1)
	LT	/lt/	Mayak Large inscr., 9th c., Sec. 7.2.9, in Khazar (2)
	OPEN M	/m/	Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (2) Mayaki Amphora, 8th-9lh c., Sec. 7.2.2, in Ogur (1) Khumara, 9 — 10,h c., Fig. 7.2.5-2b,3, in Khazar (4) Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic (2) Bilingual Khazar Coin, Sec. 7.2.11, in Khazar (1) Ermen Tolga, 8th— 10th c., Sec. 7.2.12, in Khazar, mirrored: (4) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (2) Mayak Smaller Inscr., 9th c., Fig. 7.2.10-2,, in Ogur, mirrored: (1) Mayak Smaller, Fig. 7.2.10-4, in As-Alan, mirrored: (2)
	ROUND M	/m/	Novocherkassk, 9th— 10th c., Sec. 7.2.6, in Kypchak (2)
	N	/n/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (1) Khumara, 9,h-10th c„ Fig. 7.2.5-2b, 3, in Khazar (3) Homokmegy-Halom, 10th c., Sec. 7.2.13, in Turkic (2) Mayak Large, 9th c., Sec. 7.2.9, in Khazar, partly, in lig. (5) Mayak Smaller, 9lh c., Fig. 7.2.10-3, in As-Alan, in lig. (1) Bilyar, 1 l,h-13th c„ Sec. 7.2.16, in Ogur(l)
	ANGLED N	/n/	Khumara, Fig. 7.2.5-lb,2b,3, in Khazar, glyph: (5) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (2)
	FORKED N	/n/	Achiktash, Sec. 7.2.4, in Kypchak, glyph: (2) Mayaki Amphora, 8th-9th c., Sec. 7.2.2, in Ogur, glyph: (4) Khumara, 9th— 10th c., Fig. 7.2.5-3, in Khazar, glyph: (1) Novocherkassk, 9th—10th c., Sec. 7.2.6, in Kypchak (1) Stanitsa, 9th— 10th c., Sec. 7.2.7, in As-Alan, partly in lig. (4) Mayak Smaller, 9th c., Fig. 7.2.10-1,5a,b, in Khazar (2) Mayak Smaller, 9,h c., Fig. 7.2.10-3, in As-Alan (1)
	NG	/ŋ/	Ermen Tolga, 8th — 10th c., Sec. 7.2.12, in Khazar (1) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (2)
	FORKED O	/o/u/	Khumara, 9th-10th c„ Fig. 7.2.5-3, in Khazar (1) Novocherkassk, 9th—10th c., Sec. 7.2.6, in Kypchak, glyph: (1) Village Krivyanskoe, Sec. 7.2.7, in As-Alan, glyph: (6) Mayak Large, Sec. 7.2.9, in Khazar, partly in lig.: (4) Mayak Smaller, Fig. 7.2.10-5a,b, in Khazar, in lig. (1) Kievan Letter, 955-961, Sec. 7.2.15, in Khazar (1)
	P	/p/	Achiktash, Sec. 7.2.4, in Kypchak, glyph: 3 (1)
	SIMPLE P	/p/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (2) Khumara, 9th—10th c., Fig. 7.2.5-4, in Ogur (1) Minusinsk, 8th-9,h c., Sec. 7.2.8, in Turkic (2)
	ANGLED Q	/q/uq/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak, sound: /uq/ (2) Homokmegy-Halom, Sec. 7.2.13, in Turkic, sound: /q/ (1)
	ARCHED Q	/q/	Mayak Smaller inscr., 9th c., Fig. 7.2.10-1, in Khazar (1)
	OPEN Q	/q/	Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (5)
	CLOSE Q	/q/	Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar (1) Mayak Large Inscr., 9th c., Sec. 7.2.9, in Khazar (1) Mayak Smaller, 9th c., Fig. 7.2.10-5a,b, in Khazar (1) Kievan Letter, 955-961, Sec. 7.2.15, in Khazar, in ligature (1)
	R	/r/	Mayaki Amphora, 8th-9th c., Sec. 7.2.2, in Ogur, glyph: (2) Novocherkassk, Sec. 7.2.6, in Kypchak, in ligature (2) Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar (2) Mayak Large, 9th c., Sec. 7.2.9, in Khazar, glyph: (2) Mayak Smaller, Fig. 7.2.10-2, in Ogur, glyph: (1)
	ARCHED R	/r/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak, glyph: (5) Village Krivyanskoe, 9,h— 10th c., Sec. 7.2.7, in As-Alan (1)
	CLOSE R	/r/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (1) Achiktash, first half of the 8th c., Sec. 7.2.4, in Kypchak (1) Khumara, 9th—10th c., Fig. 7.2.5-lb,2b, in Khazar (2) Khumara, 9th—10th c., Fig. 7.2.5-4, in Ogur (1) Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic (3)
	ARCHED SH	/ʃ/	Mayak Large, 9th c., Sec. 7.2.9, in Khazar (1)
	CIRCLE ENDED SH	/ʃ/	Village Krivyanskoe, 9th-10th c., Sec. 7.2.7, in As-Alan (1)
	ZS (SH)	/ʃ/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (1) Devitsa Coins, 8th-9th c., Sec. 7.2.3, in Khazar, in ligature (2) Novocherkassk, 9th-10th c., Sec. 7.2.6, in Kypchak (3) Ermen Tolga, 8th—10th c., Sec. 7.2.12, in Khazar (4)
	SHL (SHL)	/ɬ/	Bilyar, 1 l,h-13th c., Sec. 7.2.16, in Ogur (1)
	SZ (S)	/s/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar (1) Khumara, 9th-10th c„ Fig. 7.2.5-4, in Ogur (1) Homokmegy-Halom, 10th c., Sec. 7.2.13, in Turkic (2) Mayak Large Inscr., 9th c„ Sec. 7.2.9, in Khazar (2) Mayak Smaller, 9th c., Fig. 7.2.10-4, in As-Alan (1)
	FORKED SZ (S)	/s/	Minusinsk, 8th-9th c., Sec. 7.2.S, in Turkic, simpler shape: (1) Khumara, 7.2.5-2b,3, in Khazar, simpler shape (2) Krivyanskoe, 9th-10th c., Sec. 1.2.1, in As-Alan (1) Mayak Large Inscr., 9th c., Sec. 1.2.9, in Khazar (1) Mayak Smaller Inscr., 9th c., Fig. 1.2.10-2, in Ogur (1)
	ARCHED T	/t/	Khumara, 9th-10th c., Fig. 7.2.5-lb, in Khazar (1)
	CENTRAL T	/t/	Khumara, 9th-10,h c., Fig. 1.2.5-4, in Ogur (1) Krivyanskoe, 9' -10th c., Sec. 1.2.1, in As-Alan (1) Mayak Smaller Inscr., 9th c„ Fig. 1.2.10-2, in Ogur (1)
	CLOSE T	/t/	Bilyar, 11th-13th c., Sec. 1.2.16, in Ogur (1)
	OPEN T	/t/	Achiktash, 8th c., Sec. 1.2.4, in Kypchak (8) Khumara, 9th—10lh c., Fig. 1.2.5-lb,2b, in Khazar (4) Mayak Large, 9th c., Sec. 1.2.9, in Khazar, glyph: (1) Alsöszentmihâlyfalva, Sec. 1.2.14, in Khazar, in lig. (1)
	ARCHED UE	/ø/y/	Bilingual Khazar Coin, Sec. 1.2.11, in Khazar, partly in lig. (2) Ermen Tolga, 8th—10th c., Sec. 1.2.12, in Khazar, glyph: (7) Mayak Large, 9th c., Sec. 1.2.9, in Khazar, glyph: (2)
	SHARP UE	/ø/y/	Minusinsk, Sec. 1.2.8, in Turkic, glyph: N, sound: /y/ (3) Khumara, 9th—10th c., Fig. 1.2.5-lb, 2b in Khazar, sound: /ø/ (2) Alsöszentmihâlyfalva, Sec. 1.2.14, in Khazar, sound: /y/ (1)
	OPEN V	/β/, in Common Turkic at the beginning of the word: /b-/, in As-Alan: /v/	Mayaki Amphora, 8th-9th c., Sec. 1.2.2, in Ogur (1) Novocherkassk, 9th—10th c., Sec. 1.2.6, in Kypchak (1) Krivyanskoe, 9th— 10th c., Sec. 1.2.1, in As-Alan (2) Bilingual Khazar Coin, Sec. 1.2.11, in Khazar (1) Ermen Tolga, 8th— 10th c., Sec. 1.2.12, in Khazar (2) Mayak Fortress Large, 9th c., Sec. 1.2.9, in Khazar (1)
	CENTRAL Z	/z/t/	Khumara, 9th—10th c., Fig. 1.2.5-3, in Khazar (1), sound: /z/ Krivyanskoe, Sec. 1.2.1, in As-Alan (1), sound: /z/ Minusinsk, Sec. 1.2.8, in Turkic (1), glyph: sound: /t/

468 Hosszu 2011, pp. 6-8 & 34-35
142, 143, 144, 145, 146, 147

In KR relics theOPEN V denoted /β/b-/v/. In current Turkic languages, /β/ also exists. For instance, in the Turkish Latin-based orthography, the character v represents /β/ in some cases.⁴⁶⁹

7.1.3. Diacritic mark and punctuation

There is only one known Khazar Rovas diacritic mark (Table 7.1.3-1); its role is signifying the end of the word. There also exist punctuation marks in KR (Table 7.1.3-2).

469 Gökser & Kerlake 2005, pp. 6-7
147

Table 7.1.3-1: The diacritic mark of KR

Glyph	Name of the KR character	Examples of the glyph in relics
	COMBINING DOT ABOVE	Novocherkassk, 9^lh—10th c., Sec. 7.2.6, in Kypchak Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar

Table 7.1.3-2: The Khazar Rovas separators and examples of each glyph in relics

Glyph	Name of the KR character	Examples of the glyph in relics
•	WORD SEPARATOR MIDDLE DOT	Novocherkassk, 9th—10th c., Sec. 7.2.6, in Kypchak Village Krivyanskoe, 9th—10th c., Sec. 7.2.7, in As-Alan Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar
:	COLON	Mayaki Amphora, 8th-9th c., Sec. 7.2.2, in Ogur Ermen Tolga, 8th— 10th c., Sec. 7.2.12, in Khazar
	TRICOLON	Village Krivyanskoe, 9th— 10th c., Sec. 7.2.7, in As-Alan
ı	WORD SEPARATOR VERTICAL BAR	Khumara 9th -10th c., Fig. 7.2.5-1b, in Khazar
	KHAZARIAN ROVAS SEPARATOR LARGE	Ermen Tolga, 8th— 10th c., Sec. 7.2.12, in Khazar Minusinsk, Sec. 7.2.8, in Turkic with also a variant: Mayak Large, 9th c., Sec. 7.2.9, in Khazar Alsöszentmihâlyfalva, 920-952, Sec. 7.2.14, in Khazar

7.1.4. Harmonization of consonants with vowels

The Khazar Rovas consonants are occasionally harmonized with the vowels of their syllables (Table 7.1.4-1). Oppositely to the Old Turkic consonants, this harmonization is not coherent.⁴⁷⁰ A consonant is called velar if it is used near back vowels, and it is called palatal if it is used near front vowels. The back (velar) and front (palatal) vowels are listed in Table A-l. In some cases, the Khazar Rovas characters of the velar-palatal consonant pairs were even exchanged, which is a unique feature of KR. Moreover, in the Khazar Rovas inscription of the Mayaki Amphora (Sec. 7.2.2) the R /r/ was used in both velar and palatal syllables. (R is neither velar, nor palatial, i.e. neutral consonant)

Table 7.1.4-1: Examples of the consonants being harmonized and not harmonized with the vowels of their syllables

Harmonized with back (velar) vowels (called velar consonants)	Harmonized with front (palatal) vowels (called palatal consonants)	Examples
ARCHED B /b/p/	ANGLED B /b/,RAISED B /b/	Jitkov, 8th c., Sec. 7.2.1, in Khazar Mayak Large, 9th c., Sec. 7.2.9, in Khazar
ARCHED D /d/di/dɛ/j/dʒ/	SHARP D /d/ð/dʒ/	Jitkov, 8 c., Sec. 7.2.1, in Khazar Mayak Large, 9th c., Sec. 7.2.9, in Khazar
, N GH /ɣ/	CLOSE G /g/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic Mayak Large, 9th c., Sec. 7.2.9, in Khazar
N GH /ɣ/	FORKED G /g/	Khumara, ca. 9th c., Fig. 7.2.5-4, in Ogur
N /n/	FORKED N /n/,ANGLED N /n/	Khumara, ca. 9th c., Sec. 7.2.5, in Khazar Mayak Large, 9th c., Sec. 7.2.9, in Khazar
CLOSE R /r/	ARCHED R /r/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak
(There is no occurrence in relics)	CLOSE R /r/	Minusinsk, 8th-9th c., Sec. 7.2.8, in Turkic
(There is no occurrence in relics)	, R /r/	Mayak Smaller inscription, 9th c. Fig. 7.2.10-2, in Ogur Alsöszentmihâlyfalva, Sec. 7.2.14, in Khazar
R/r/	R/r/	Mayaki, 8th-9th c., Sec. 7.2.2, in Ogur
OPEN T /t/	ARCHED T /t/	Khumara, ca. 9th c., Sec. 7.2.5, in Khazar
(There is no occurrence in relics)	, OPEN T/t/	Achiktash, Fig. 7.2.4-1, in Kypchak Minusinsk, Fig. 7.2.8-1, in Turkic Mayak Large, Fig. 7.2.9-2, in Khazar

470 Vekony 2004a, p. 193
148, 149

7.1.5. Ligatures

In Khazar Rovas, ligatures were frequently used. They are only stylistic, similarly to CBR ligatures (Sec. 6.1.5). However, two ligatures are systematic and used regularly (Table 7.1.5-1): DA and DI. The sound values of DI are /de/, /di/, and /ed/. Surely, its constituents areARCHED D /d/ andANGLED I. This supports that theANGLED I originally represented both /e/ and /i/, despite of the fact that in the surviving KR relics, theANGLED I represented /i/ only. It is noteworthy that in the Old Turkic script, which is also a descendant of the Early Steppe script, similarly to KR, the Old Turkic I represented both /e/ and /i/ (Sec. 4.1).

Table 7.1.5-1: The systematic Khazar Rovas ligatures

Glyph	Name	Transcription	Constituents	Examples of the glyph in relics
	DA	/da/	ARCHED D /d/ + Y FORKED A /a/	Krivyanskoe, Sec. 7.2.7, in As-Alan, sound: /da/ Mayak Large, Sec. 7.2.9, in Khazar, sound: /da/ Mayak Smaller, Fig. 7.2.10-1, in Khazar, /da/
	DI	/de/di/ed/	ARCHED D /d/ + ANGLED I /i/e/	Humara, Fig. 7.2.5-4, in Ogur, sound: /di/ Alsöszentmihâlyfalva, Sec. 7.2.14, in Khazar, /ed/ Ermen Tolga, Sec. 7.2.12, in Khazar, sound: /de/ Mayak Large, Sec. 7.2.9, in Khazar, sound: /de/

149

Table 7.1.5-2 presents the non-systematic Khazar Rovas ligatures being applied on occasion.

Name	Transcription	Constituents	Examples of the glyph in relics
AA	/ʃi/, /iʃ/, /aa/	FORKED E /a/ +FORKED E /a/	Village Krivyanskoe, Sec. 1.2.1, in As-Alan
AN	/an/	ARCHED E /a/ + N /n/	Khumara, 9th—10th c., Fig. 1.2.5-3, in Khazar
ATLIGH	/^ətlⁱɣ/	OPEN T /t/ +SIMPLE L /l/ + N GH /ɣ/	Alsöszentmihâlyfalva, 920-952, Sec. 1.2.14, in Khazar
BAN	/ban/	RAISED B /b/ + Y FORKED A /a/ + N /n/	Mayak Smaller inscription, 9th c., Fig. 1.2.10-3, in As-Alan
CSI (SHI)	/tʃi/	TRIPLE CS (SH/S) /tʃ/ + ANGLED I /i/	Khumara, 9th- 10th c., Fig. 1.2.5-4, in Ogur
DKE	/ðkɛ/	SHARP D /ð/ + FORKED K /k/ + ARCHED E /ɛ/	Jitkov, first third of 8th c., Sec. 7.2.1, in Khazar
ER	/ɛr/	FORKED E /ɛ/ + R /r/	Novocherkassk, 9th—10th c., Sec. 1.2.6, in Kypchak
IL	/il/	ANGLED I /i/ (mirrored glyph) + SIMPLE L /l/	Kievan Letter, 955-961, Sec. 1.2.15, in Khazar
IN	/ïn/	ARCHED I /ï/ + FORKED N /n/	Khumara, ca. 9th c., Sec. 1.2.5, in Khazar
IQ	/ïq/	ANGLED I /i/ (mirrored glyph) + CLOSE Q /q/	Kievan Letter, 955-961, Sec. 1.2.15, in Khazar
JI	/ji/	CLOSE J /j/ + ANGLED I /i/ (mirrored glyph)	Mayak Smaller inscription, 9th c., Fig. 1.2.10-4, in As-Alan
ISHAD	/ⁱʃ^að/	ZS (SH) /ʃ/ + SHARP D /ð/	Devitsa Coins, c., Sec. 1.2.3, in Khazar
NI	/ni/	FORKED N /n/ + ARCHED I /i/	Krivyanskoe, 9th— 10th c., Sec. 1.2.7, in As-Alan, ornamental
NU 1	/nu/on/	FORKED N /n/ + FORKED O /u/	Krivyanskoe, Sec. 1.2.1, in As-Alan Mayak Smaller inscription, 9th c., Fig. 1.2.10-5a,b, in Khazar
NU 2	/nu/	N /n/ + FORKED O /u/	Mayak Large, 9th c., Sec. 1.2.9, in Khazar
ON	/on/	FORKED O /u/ + N /n/	Mayak Large, 9th c., Sec. 1.2.9, in Khazar
RAT	/rat/	CLOSE R /r/ +ARCHED E /ɛ/ + OPEN T /t/	Khumara, ca. 9th c., Sec. 1.2.5, in Khazar
SI	/sï/	SZ (S) /s/ +ANGLED I /ï/	Achiktash, 8th c., Sec. 1.2.4, in Kypchak
SHID	/ʃið/	ZS (SH) /ʃ/ +ANGLED I /i/ +SHARP D /ð/	Achiktash, 8th c., Sec. 7.2.4, in Kypchak
TEGH	/t^eɣ/	OPEN T /t/ + N GH /ɣ/	Khumara, ca. 9th c., Sec. 7.2.5, in Khazar
UED	/øð/	ARCHED UE /ø/ +SHARP D /ð/	Bilingual Khazar Coin, Sec. 7.2.11, in Khazar

150, 151

7.2. Relics of the Khazar Rovas

Map 7.2-1 presents the locations of KR archaeological finds in Eurasia. Most of the Khazar Rovas relics were found in the territory of the late Khazar Kaganate, north of the Caucasus.

Map 7.2-1: Locations of the Khazar Rovas archaeological finds in Eurasia excluding the relics of the Carpathian Basin

7.2.1. Bow Cover of Jitkov

In 1986, Y. I. Bespalyi found a horn cover for a bow with a Rovas inscription in burial No. 1 of kurgan No. 4 on the field Jitkov II in the Rostov region.⁴⁷¹ The relic is dated by some coin finds. Among them, there is a dirham from Nahr Tira dated to 716-717, and a solidus of Leo III (ruled in 717-720). These coins date the burial to the first third of the 8^lh century.⁴⁷²

471 Semenov 1988, p. 109
472 TürkicWorld, web site
151

Figure 7.2.1-1: The inscription on the bow cover found at Jitkov, first third of the 8th century⁴⁷³

Table 7.2.1-1: Transcription of the Jitkov Inscription, first third of the 8th century

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar)	Thkeib a xo ɣiyab ɣibe thib apaʒ n:aaraspa ɣaʒ shithi
Translation from Common Turkic (Khazar)	‘We send Apsar Jap [the] bow. Shoot a mansion, wealth, oh to us!’

The ligature DKE /ðkɛ/ is composed of the followings: SHARP D /ð/ + FORKED K /k/ + ARCHED E /ɛ/.

7.2.2. Mayaki Amphora Inscription in Ogur

The Mayaki Amphora Inscription was found in the motte-and-bailey (citadel) of Mayaki, on the right bank of the Donets River. Its drawings were created by I. L. Kyzlasov (Fig. 7.2.2-1 and 2). The amphora was made in the 8 - 9th centuries, when the fortress was used.

Figure 7.2.2-1: The Mayaki Amphora Inscription, 8th-9th centuries⁴⁷⁴

Figure 7.2.2-2: Another drawing of the Mayaki Amphora Inscription⁴⁷⁵ On this relic, the glyph R /r/ - a variant of the R /r/ was used.

Table 7.2.2-1: The transcription of the inscription on the Mayaki Amphora

Written with normalized Khazar Rovas font	[unreadable]
IPA phonetic transcription
Common Turkic (Khazar)	Nob merʒ nï naɣax rana
Translation from Ogur	‘(This is) of Kagan Nar [or Anar], Twenty won.'

⁴⁷³    Semenov 1988, p. 108
⁴⁷⁴    Klyashtomy 1979
⁴⁷⁵    Klyashtomy 1979
152

The Ogur word won was a measurement unit. The first two characters of the inscription were damaged; their transcription presumably is bu. The name Nar (or Anar) was an ancient Turkic personal name.⁴⁷⁶ Since the transition /q/ > /x/ had been completed by the beginning of the 10th century in Ogur,⁴⁷⁷ the use of the word /xaɣan/ instead of the /qaɣan/ is explicable in the inscription.

The Oguz tribes moved into the Aral neighborhood at about 750, absorbing a spectrum of the local tribes known under a collective name Alans (Field or Steppe People, Polyans in Slavic) who were big on /h/ (/x/), it can be expected that both the local version /xaɣan/ and the Oguz version /qaɣan/ remained in circulation through the 10th c. and way beyond. The Arab (Moslem) writers used the form /xaɣan/, because that form was used by the southwestern tribes they first encountered, or that was an Arabicized form due to the specifics of the Arab phonetics. Either way, a quasi-scientific presumption on codified single Oguz language, and the assertion that the blend of the local and migrant tribes spoke a common Oguz language without dialectal differences, are pretentious fallacies. The -h- dialect of the Aral basin followed Alans across Europe to Africa, and is prominant in the New World under a name Castilian (Castellano) of the Argentinian Spanish.

7.2.3. The Devitsa Coins

In 1939, the Devitsa (also called Devica) treasure was found near the Don River. It contains 237 coins. According to Bykov, these coins are Khazar products ⁴⁷⁸ and their origin is between 754 and 811/812.⁴⁷⁹ This opinion is supported by scholars, especially Ludwig⁴⁸⁰ and Bâlint⁴⁸¹

On several coins, the same ligature can be observed (Table 7.2.3-1)⁴⁸² It means Ishad /iʃað/, the alternate name of the vice-king Ilik (i.e. Ulug, aka Yilig, a short for Ulug-Bek, aka Bek, aka Kagan-Bek, a title of a Prime Minister at a presiding Kagan) of the Khazar Kaganate.⁴⁸³ This inscription authenticated the coins. According to Golden, Ishad (a Göktürk military rank used among southeastern tribes) is a version of an Iranian (Persian?) dignitary name (title?).⁴⁸⁴ Based on these finds, the word Ishad was used in the 8th- early 9th centuries. However, the dignitary name Ilik was used later, in the 10th century. Besides Ishad and Ilik, there was a third name of the vice-king, namely /bɛk/; its other form was /beg/.⁴⁸⁵

Table 7.2.3-1: The ligature of the Devitsa Coins from the 8th-9^lh centuries and its transcription

Ligature on the Devitsa Coins (glyph variations)
The elements of the ligature written with normalized Khazar Rovas font	(SH) /ʃ/ (I) /i/
IPA phonetic transcription	/ⁱʃ^að/
Common Turkic (Khazar)	Ishath
Translation from Common Turkic (Khazar)	‘Ishad’

Alternatively, the “ligature” was suggested to be a tamga of the ruler, in the Devitsa Hoard of the Kagan Bulan (760-805), whose Beks were in sequence Bulan Sabriel (ca 740) and Obadia (ca 786-809). A name or a title-name of a subordinate minted on the coin would indicate a claim to suzerinty and treated as a rebellion, an unlikely situation with two consequtive Beks under the sama Kagan. Coin images (Fig. 21) and extracts from the table (Table 3) are cited from A. Mukhamadiev, 2005, Ancient Coins of Kazan Kazan, Tatar Publishing house, ISBN 5-298-04057-8.

Images of the Khazar coins
Fig. 21. Khazar dirham	Khazar coin - immitation

	Khazar coin with Türkic runiform inscription Found in Kaluga region (Central Russia)

	Türkic runes in the legend of this coin are not Orkhon or Enisei alphabet, but definitely from Western areas (Southern Russia, Volga region, Northern Caucasus). Attribution: Itil Bulgar or Khazar Kaganate (?), 9-10th c. (?)

Table 3
Khazar imitations (Devitsky hoard)

Coin type	Coin dates	Dating mark on reverse	Reign years of rulers and existence of symbols		Qty of coins	Kagan and Bek	Tamga
VIII	150/767-68	Symbol of creed in two circles	136-158	754-775	1	Bulan (760-805)/Bulan Sabriel (ca 740)
IX	150/767-68	Ditto	136-158	754-775	1	Bulan (760-805)/Bulan Sabriel (ca 740)
X	150/767-68	al-Mahdi as heir	145-158	762-63-775	1	Bulan (760-805)/Bulan Sabriel (ca 740)
XI	150/767-68	Zur-p-riyasatain (Fadl ibn Sahl)	196—205	811-12-820-21	38	Bulan (760-805)/Bulan Sabriel (ca 740)
XII	162/778-79	Symbol of creed in two circles	136-158	754-775	1	Bulan (760-805)/Obadia (ca 786-809)
	837/838	Symbol of creed Musa rasul Allah “Moses is the Messenger of God”			5		—

476 Vekony 2004a, p. 272
477 Rona-Tas 1982b, pp. 163-164
478 Bykov 1971, pp. 26-35
479 Bykov 1971, pp. 65-66
480 Ludwig 1982, p. 274
481 Bâlint 1980, p. 384; Bâlint 1981. p. 410
482 Shake 2000, p. 35; Kovalev 2004, p. 113
483 Vekony 1997a, p. 28
484 Golden 1980, pp. 206 208
485 Golb & Pritsak 1982
153

7.2.4. The Achiktash Inscription from the Talas Valley in Kypchak

Fig. 7.2.4-1 presents the Achiktash Inscription (Wooden Stick of Talas) from the first half of the 8th century.

See Codex_EuroAsiaticAchiktash_En.htm for details. The stick, buried under 5 m of soil in a well of an old mine, should be assosiated with the mining population of the time, which for millennia is anchored down by the deposits, and lives through all regimes. The mine bore tests showed 57% Fe, 4% Cu, and 1% S, hence look for the iron-smith locals right on the main trace of the Silk Road (Ian Blanchard, 2001, Mining, Metallurgy and Minting in the Middle Ages: Asiatic supremacy, Franz Steiner Verlag, ISBN 9783515079587). The association with Kipchaks is more than dubious, but the Saka Piedmonter tribes of the Türks, known as iron-smiths from the Chinese annals, are quite suitable (till the C¹⁴ analysis of the stick and the mine). The geographically nearest runiform inscriptions are the Issyk inscription (4th c. BC) and Kultobe inscribed ceramic bricks of standard Middle Asian type 27x27x6 cm (4th-1st cc. BC).

Issyk

Kultobe

Figure 7.2.4-1: The Achiktash Inscription (first half of the 8th c.)⁴⁸⁶

The inscription is on the four sides of a wooden stick. The stick was broken into three parts, but the middle one was never found. Therefore, the middle part of the text in each row is missing; the transcription is shown in Table 7.2.4-1 ⁴⁸⁷ According to Vekony, the occasion for creating the inscription was probably the accession of a new ruler to the throne.⁴⁸⁸

In the inscription, the glyph - a variant of the ARCHED R /r/ and the glyph - a mirrored variant of the FORKED N /n/ were used. The glyph was normalized as P /p/.

Table 7.2.4-1: Transcription of the Achiktash Inscription

Written with normalized Khazar Rovas font
(right-to-left)

IPA phonetic transcription
(left-to-right)

Common Turkic (Khazar) transcription
(left-to-right)

Tir ïthuq ethe ørin...
Qathɣuq hide itthïq oqïmïn mith....
Tegil aj suht osyt...
Hith qasar uthusï

bïtïrh[ï(ɛ)ngïth]
ishihtibith qop
ɛrtɛr t[anïng]

Translation from the Kypchak

‘He says: the throne of the holy dominus you write [singular or plural] We ... reading the written stick sent by you we all heard it Ay! Say it: Sogdian vengeance ... avoid [you], [you] k[now] You, Khazars. Good (= true, credible). End.’

Ay was a personal name meaning ‘Moon’ in Proto-Turkic (what “Proto-Turkic” means?). In the fourth row, the name Khazar is found: /qasar/ ‘Khazar’ and the name of the Sogdians is in the third row: /s^uht/ ‘Sogdian’. Vekony supposed that in this inscription, the KR TRIPLE CS (SH/S) /tʃ/ represents /s/.⁴⁸⁹ This is possible, if the /tʃ/ > /s/ change happened in the language of this inscription. For instance, such a transition occurred in the Bashkir language.⁴⁹⁰

⁴⁸⁶    Vekony 2004a, p. 287
⁴⁸⁷    Vekony 1986, Fig. 1
⁴⁸⁸    Vekony 2004a, p. 293
⁴⁸⁹    Vekony 2004a, p. 294
⁴⁹⁰    Vâsâry 2010-2011
154

7.2.5. The Khumara building inscriptions

Khumara is an old Khazar fortress on the right hand side of the Kuban River, currently in the Karachay-Cherkess Republic (Russia). The Khumara Fortress was near to a very important road, which was the main artery between Khwarezm and Byzantium in the Early Medieval Times.⁴⁹¹ The total area of the huge fortress was a quarter km², the total length of its walls reached 1900 meters. A.V. Gadlo showed that the fortress was built in the beginning of the 8th century.⁴⁹² Fig 7.2.5-1 a presents the photograph of a building inscription on a sandstone block found in 1962, and Fig.

7.2.5-lb shows its drawing. The original stone was re-used for the construction of the fortification wall and it was broken into two pieces. The origin of the inscription is between the mid-9^,h century and beginning of the 10th century.⁴⁹³

Figure 7.2.5-la: Photograph of the One-row Inscription on a fragment of a sandstone block, between middle of the 9th and 10th centuries⁴⁹⁴

Figure 7.2.5-lb: Drawing of the One-row Inscription⁴⁹⁵

Table 7.2.5-1: Transcription of the One-row Khazar Rovas Inscription of Khumara

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	øth tegin daratï
Translation from Common Turkic (Khazar)	‘Öd Tegin made [it].’

The word /d^arati/ is the ancestor of the Hungarian word gyârt /da:rt/ ‘produce, make’. The Öz /øz/ is a well-known Turkic personal name.⁴⁹⁶ According to Vekony, the transition /z/ > /ð/ (/th/) happened in Khazar. Therefore, Öd /øð/ (/öth/) was identical with Öz /øz/. The /tegin/ (Prince) was a Khazar title (dignitary name). There is an example for this dignitary name in the sources: the name “Er Tegin” from the 680s.⁴⁹⁷ In the Western Turkic Kaganate, the title tegin was also used, as it is attested to in inscriptions found in East Iran.⁴⁹⁸ The word tegin survived as a personal name in the 10th century: a leader of the Magyars was called Teteny (its other form: Tühütüm).⁴⁹⁹ This name can be found in the name of Teteny Village (present-day Tadten in Burgenland, Austria) near the Lake Ferto, in the name of Tattendorf 'Tatten Village’ (present-day east of Vienna, Austria) near the Triesting River, and in the name of Budapest District 22: Nagyteteny ‘Great Teteny’.⁵⁰⁰

The speculation on /z/ > /ð/ (/th/) transition conflicts with attestation that interdental th was widespread in the Old Türkic, depicted as δ in transcriptions, and in the old Türkic Horesmian, depicted as and identical to the Futhark Þþ. Likely, the transition went in opposite way, the alien speakers were adopting the Türkic th in a veriety of approximations, including /z/, /t/, /d/, and /f/.

⁴⁹¹    Erdelyi 1983b, p. 264
⁴⁹²    Gadlo 1979, pp. 152-153
⁴⁹³    Baichorov 1989, p. 67, pp. 170-173; Türkic World, web site
⁴⁹⁴    Kuznetsov 1963, Fig. 2 -2
⁴⁹⁵    Vekony 1987a, p. 27; Vekony 2004b
⁴⁹⁶    Nadeljaev et al (ed.) 1969, p. 395
⁴⁹⁷    Golden 1980, pp. 186-187
⁴⁹⁸    Harmatta 1994, pp. 149-164
⁴⁹⁹    Györffy 1998, p. 135
155

Figure 7.2.5-2a: Photograph of Three-row Inscription from the Khumara fortress⁵⁰¹

The meaning of the One-row Inscription in Fig. 7.2.5-1a matches the first row of the inscription of the Fig. 7.2.5-2a. In the first row of Fig. 7.2.5-2a, the second symbol on the left is a ligature of the characters CLOSE R /r/,ARCHED E /a/, and OPEN T /t/. These characters were used as individual glyphs in Fig. 7.2.5-la. In the third row, the ligature contains OPEN T /t/ and N GH /ɣ/.

In the word /t^eɣⁱne/ there are front vowels. However, in other parts of this Three-row Inscription, the CLOSE G /g/ was applied near front vowels and another character N GH /ɣ/ is used near back vowels. The probable reason of this inconsistency is thatOPEN T /t/ and N GH /ɣ/ was easier to apply to create the ligature . Another example for altering the velar and palatal consonants is the use of the velarOPEN T /t/ in the first row instead of the palatalARCHED T /t/. It is noteworthy that in this case theOPEN T /t/ was not used in ligature. To summarize, this inscription demonstrates the inconsistent use of the velar and palatal consonants in KR (Sect. 7.1.4).

A tendency to blame scribes and carvers is a longstanding philological tradition. A lazy carver may not be a culprit, though. Most of the time, a palatal syllable carries a different meaning from the velar syllable, and to confuse them is akin to replacing red stripes with green stripes on a national flag, that completely changes the message. A few words have two versions, one with palatal phonetics, and the other with velar phonetics. Frequently, these few words also contain neutral syllables, and can be swung either way. Absent that, different meanings should theoretically be suspected for tegine /t^egⁱne/ and täkïnä /t^ək^ïnə/ before invoking a lazy carver. In the end, the linguistic and phonetical attributions of the carvings and the use of ligatures are conditional on what is not known.

Figure 7.2.5-2b: Drawing of the Three-row Inscription⁵⁰²

500 Bartha 1984, p. 577; Györtly 1998, p. 135; Zelliger 2010-2011
501 Kuznetsov 1963, Fig. 2-2
502 Vekony 2004b
156

Table 7.2.5-2: Transcription of the Three-row Inscription of Khumara

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	ød tegin daratï thïnïmïthïɣ asman øthyg teɣine aj
Translation from Common Turkic (Khazar)	‘Öd Tegin (had it) made our nest [=home], As(es). Call it Ödüg Tegin.'

Table 7.2.5-2: Transcription of the Three-row Inscription of Khumara

The drawing in Fig. 7.2.5-3 was made by the student A.D. Besleneev on his summer holiday; hence, its accuracy is possibly limited. There is no photograph of the original relic, since it was destroyed before the arrival of the archaeologists.⁵⁰³ In the inscription, the glyph 0 is a variant of D CLOSE J /j/. The word /aj/ means ‘speak, say, call’ in Turkic. The word /*asman/ probably mean ‘As’ or ‘Ases’, since /as/ was the name of the As-Alans and /man/ is a Turkic affix emphasizing the meaning (nation or tribe name). (It could also mean various types of rams, like a castrated ram; the idea on Ases comes from Ases being a tribe of what collectively was called Alan “Steppe-men, Flatlander”; the modern reference to Ases may be too politicized)

Text is rich in information: the Khazar fortress of Khumara was built by Öd Tegin; therefore, it was named Ödüg Tegin. In the name of Ödüg, the /g/ is the voiced version of the Turkic diminutive suffix /k/. There are other examples for this structure: personal name + title > geographical name. For instance, Phanagoria (the ancient Greek colony on the Taman peninsula between the Black Sea and the Sea of Azov) has an alternative name: /tamyantarqan/, in which the /tarqan/ is a title (dignitary name).

Figure 7.2.5-3: Drawing of the Two-row Inscription from Khumara Fortress⁵⁰⁵

Table 7.2.5-3: Transcription of the Two-row Inscription of Khumara

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	ïnumuzuɣ, asman øththün ebine aj
Translation from Common Turkic (Khazar)	‘Our nest [=home], As(es), call it mansion of Ödün.'

The inscription in Fig. 7.2.5-3 equals to the second and third rows of the inscription of the Table 7.2.5-2. However, there are dialectical differences between the Table 7.2.5-2 and 3. For instance, in the first row in Table 7.2.5-3 CENTRAL Z /z/ was used instead of SHARP D /ð/ in Table 7.2.5-2. A similar dialectical difference is between /ïn/ in Table 7.2.5-3 and /ð^ïn/ in Table 7.2.5-2.

503 Kuznetsov 1963, Fig. 3-1; TürkicWorld, web site
504 Ligeti 1958. p. 448; Ligeti 1977, p. 78, Kononov 1980, p. 96
505 Vekony 2004b
157

The ligature was composed of ARCHED I /ï/ and N /n/, the auxiliary slanted line was presumably applied in order to differentiate it from ) ARCHED E /a/. The /n/ in the word Ödün was probably a diminutive suffix. (There is something alarming about similarity of the name Ödün of the Caucasus Ases and the name Odin of the Scandinavian Ases, both of the princely variety)

The inscription in Fig. 7.2.5-4 is fragmentary; its transcription needed a reconstruction (drawing of the archaeologist G. Vekony based on A.M. Shcherbak’s publication).⁵⁰⁶

Figure 7.2.5-4. Drawing of a fragmentary inscription from Khumara Fortress.⁵⁰⁷
On this relic, the glyph - a mirrored variant of the N GH /ɣ/ was used.

Table 7.2.5-4: Transcription of the One-row Ogur Inscription of Khumara

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	tuɣalas, ʒapdi, bishigi ïr
Translation from Ogur	‘Tughalas made (built?), wrote its inscription.’ or: ‘N. Tughalas made (built?), wrote its inscription.’

The word /t^uɣ^al^as/ is an ethnic name according to Ibn Rustah.⁵⁰⁸ The last word /ïr/ is the Ogur version of the Common Turkic verb /yaz/ ‘draw’.⁵⁰⁹ The evidence of the existence of the Ogur verb /ïr/ is the Hungarian verb ir /i:r/ ‘write’.⁵¹⁰ The verb /ʒ^apdi/ (< Proto-Turkic /yap/) means ‘built, did’, the KR SHARP D is transcribed with /ʒ/, supposing (suggesting) that the inscription is in Ogur. The word /bⁱtʃig/ is the Ogur version of the Common Turkic /bitig/.⁵¹¹ Since the Tughalas /t^uɣ^al^as/ can be an ethnic name, it could also be a personal name. Moreover, since the beginning of the inscription is broken, the “<Personal name>, the Tughalas” structure is also a possible interpretation. An uncertainty of the transcription is the ligature /di/ in the reconstruction of the word > /dʒ^apdi/ ‘made’, since the lower part of the ligature is damaged.

The described specifics of the Ogur vs. Oguz (aka Common Turkic) variety suggests that the Oguz form with a prosthetic semi-consonant is an Ogur form /yaz/, adopted into Oguz languages with the anlaut semi-consonant and preserved there, while the form /ïr/ in the inscription reflects the loss of the initial semi-consonant typical for the Oguz languages. With the north/south divide between the Oguz and Ogur, that would indicate that the Türkic term “write” originated in the south-west. The verb /ʒ^ap-/ would be a conforming Ogur form with anlaut prosthetic consonant for the Oguz counterpart with anlaut semi-consonant /yap/ that has nothing to do with the mythical “Proto-Turkic” language. The /t/ <=> /tʃ/ is a non-specific dialectal alteration, it arises under an influence of palatalized speech, like Chuvash and Slavic.

The ethnonym Taulas is still active, Taulases occupy the summit area at the Daryal Pass on its southern side. The tau/tag/taɣ/tug stands for “mountain”, -l- is an adjectival suffix a la “mountainous”, and -as is the ethnonym “As”, i.e. “Mountain As”; the reconstructed t(a)uɣ[a]l + as does not need the suggested bracketed [a], i.e Tauglas. That a Taulas was noted in inscription is natural for a migrant, everybody else on the construction crew was not from the Taulases. The modern Taulases can't be confused with the historical Taulases. The modern Taulases are predominantly descendents of the local Adyge population, of G2 Y-DNA variety (G2a1a1), that migrated from the Old Europe 4000 ybp, their Adyge-type language is now the Ossetian language. The small admixture of R1a and R1b haplogroups are the relicts of the historical As population that gave its name to the modern Taulases.

7.2.6. Novocherkassk Clay Flask Inscription in Kypchak

A clay flask with two handles was found during the construction of Salsk-Tsaritsyn (currently Volgograd) railroad in 1909.⁵¹² There are similar flask finds from around Novocherkassk with typical dimensions: width is no larger than 8 cm; diameter is about 30-35 cm. According to Artamonov, the origin of the clay flask on the Fig. 7.2.6-1 is the 9th - 10th centuries. Under the right handle, on the convex side of the flask, there is a KR inscription. Fig. 7.2.6-1 shows its photograph, and Fig. 7.2.6-2 presents its drawing.

506 Shcherbak 1962, Fig. 1
507    Vekony 2004b
508    DeGoeje, M. J. fed.) 1892
509    Vâsâry 2010-2011
510    Pallö 1982. p. 111-113
511 Röna-Tas 1992, pp. 12-13
512 Artamonov 1954, p. 264
158

Figure 7.2.6-1: The Novocherkassk Inscription (9th-10th centuries)⁵¹³

Figure 7.2.6-1: The drawing of the Novocherkassk Inscription.⁵¹⁴On this relic, the glyph - a variant of theFORKED O /o/u/ was used.

Table 7.2.6-1: Transcription of the Novocherkassk Inscription.
The transcription of the archaeologist-historian G. Vekony⁵¹⁵ was modified by Turkologist I. Vâsâry in 2011.⁵¹⁶

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Kypchak) transcription (left-to-right)	aishermith bith ishermith boshadin
Translation from Common Turkic (Kypchak)	‘Oh. Let us drink. We drink from [this] bosha.’

The word bosha was the name of a fermented alcoholic drink; Sec. 6.2.9 presents the genealogy of the word. In this case, theOPEN V probably represented /b-/, since the inscription was written in Common Turkic, where in the beginning of the word /b-/ was pronounced instead of /β-/.⁵¹⁷ The word /-dïn/ is a Turkic ablative suffix, meaning ‘from’.

7.2.7. Village Krivyanskoe Clay Flask Inscription in Preossetic (As-Alan)

In May 1942, a clay flask was discovered, at a depth of 1.5 meters, near Village Krivyanskoe (Fig. 7.2.7-1).⁵¹⁸ Its inscription was published by Artanomov in 1954, with the original punctuation by Turchaninov in 1964 and in 1971.⁵¹⁹ Figure 7.2.7-2 shows Turchaninov’s drawing.

513    Baichorov 1989, Table 113
514    Vekony 2004a, p. 243
515    Vekony 2004a, pp. 243-252
516    Vâsâry 2010-2011
517    Vâsâry 2010-2011
518    Artamonov 1954, p. 263-268
519    Turchaninov 1964 & 1971
159

Figure 7.2.7-1: The photographs of the Village Krivyanskoe Inscription (9th—10th centuries)⁵²⁰

Figure 7.2.7-2: The drawing of the Village Krivyanskoe Inscription.⁵²¹On this relic, the glyph -a variant of theFORKED O /o/u/ was used.

Table 7.2.7-1: Transcription of the Village Krivyanskoe Inscription

Written with normalized KR font
IPA phonetic transcription
Common Turkic (Kypchak?) transcription (left-to-right)	voshudanu thabu danu charith niuazag thaxichan uacheg bafistaa
Translation from As-Alan (Preossetic)	‘[This] voshu drink is yours. [This] drink should keep alive [the] one who drinks [it]. For you, Uachag wrote this.’

In this relic, the form voshu of the drink boza (English booze) occurs;⁵²² Sec. 6.2.9 presented the genealogy of this word. The sound /a/ was a Preossetic common determiner, meaning ‘this’. This /a/ is written in the duplication ofFORKED E /a/ ( /a-a/ at the end of the row, where the other /a/ is the last vowel of the word /b^afⁱsta/. In the inscription at the end of the word /ni^uz^ag/, there is the diacritic mark o COMBINING DOT ABOVE, a specific Khazar Rovas word separator. The use of this separator in the inscription was probably occasionally mixed with the other Khazar Rovas separator, the TRICOLON.

What is “Preossetic” one God knows. Probably it is a reverse projection of a blend of the modern Ossetic languages, then it is a mixture of the Adyge (Y-DNA G2a1a1) and Near Eastern (Y-DNA J2) substrate of the Ossetic, impregnated with few Turkisms accumulated for the past 1700 years. If it is an allusion to the Türkic language of Ases or Alans, or the much later Kipchaks, the enigma just grows bigger. The very term used, “Preossetic”, attests that the language has nothing to do with the Khazars and their multi-variegated Türkic ilk, including the Ases or Alans.

7.2.8. Spindle Disk of Minusinsk

In 1948, a rounded stone of a spindle was explored during the archaeological excavations of V.P. Levasova near Minusinsk. The relic is from the 8th - 9th centuries, currently preserved in the Minusinsk Museum.⁵²³ There are Old Turkic symbols on the flat surfaces of the spindle disk, and there is a Khazar Rovas inscription on its arched lateral surface. Fig. 7.2.8-1 presents the photograph of the spindle disk,⁵²⁴ and Fig. 7.2.8-2 shows the drawing and the transcription made by Vekony.⁵²⁵ The transcription was improved by Vâsâry.⁵²⁶

520   Baichorov 1989, Table 116
521   Turchaninov 1964, pp. 83 -84 & 1971, pp. 75-77
522   Vekony 2004a, p. 257
523   Vasilev 1983, p. 40, No. E-87
524   TÜRIK BITIG, web site
525   Vekony 2004b
526   Vâsâry 2010-2011
160

Figure 7.2.8-1: The photograph of the spindle disk of Minusinsk

Figure 7.2.8-2: The copy of the inscription broken into two rows

On this relic, the glyph variant of theCIRCLE ENDED I /i/ï/, of theSHARP UE /y/, of theCENTRAL Z /z/t/, and the mirrored glyph of the N GH /ɣ/ were used. The glyph - a variant of the symbol KHAZAR ROVAS SEPARATOR LARGE was also used on this relic.

Table 7.2.8-1: Transcription of the Inscription of Minusinsk.

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Kypchak) transcription (left-to-right)	ipjig jygyrryg ørtylmeth (ürtülmiş(le:rke:) Clauson p. 209) (a)suɣapam
Translation from Turkic	‘(This) thread-spindle [is] fast. My aunt (A)sugh does not hide.’

An interesting specialty of this relic is that the fifth character from the right and the last (leftmost) character of the first row in Fig. 7.2.8-2 are corrected: the writer earlier wrote TRIPLE GH /ɣ/ in both cases, and then corrected them withCLOSE G /g/. In other words, the writer first used the fricative /ɣ/, and then it was corrected to the plosive /g/. The personal name in the inscription can be either Asugh /asuɣ/ or Sugh /suɣ/. (örtül-/ürtül- is passive of “conceal, cover, hide” > “concealed by my Asug-apa”)
161

7.2.9. Mayak Fortress Large Building Inscription

In 1978, during the Soviet-Hungarian-Bulgarian common excavations of the Mayak Fortress (‘Stone Castle of Mayak’, other name: Mayaki citadel), S.A. Pletneva and G.E. Afanasyev found the longest known KR inscription.⁵²⁷ The citadel was built in the 9th century; the city itself existed from the end of the 8th century and it belonged to the territory of the As-Alans.⁵²⁸. Fig. 7.2.9-1 a and b are drawings of the architect B. Erdelyi, a participant in the excavation.⁵²⁹

Figure 7.2.9-la, b: Drawings of the Inscription of Mayak Fortress by architect B. Erdelyi.
On this relic, the glyph variants of theFORKED O /o/u/, of theDIAGONAL E /ɛ/, and of the ARCHED UE /ø/y/ were used.

a)	b)

527    Erdelyi 2004a, p. 76
528    Erdelyi 1991
529    Kyzlasov 1990; p. 16
162

Figure 7.2.9-2: The drawing of the Inscription of Mayak Fortress reconstructed by archaeologist G. Vekony

The ends of the first and second rows may show characters of a different script; those were probably engraved later.

Table 7.2.9-1 presents the transcription of the inscription.⁵³⁰

Table 7.2.9-1: Transcription of the Large Building Inscription of Mayak Fortress

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	danu aq öthyg sub asil echim xaɣan bithing deribith alti saβïrïng [...] uth dithy on munda ilte ygysh isig bedithedi ol dathu ineg [...] onaɣ tegin ebi ög
Translation from Common Turkic (Khazar)	First row: 'Don [River], white Ödüg River, As Country, [oh] my brother-khagan, our land of the Six-Savirs ...’ Second row: ‘Ud Didü On, here, in [this] country, ornamented several works. This inscription made by Ineg...’ Third row: ‘Mansion of Onagh Tegin. Extol it.’

First row: Danu is the Don River in the As-Alan language. The Aq means white; this is probably the attribute of the following river-name Ödüg or the name of another river. Ödüg is the Khazar name of the Dnieper River.⁵³¹ The /alti saβïr/ ‘Six-Savirs’ is the name of a nation, as in the Turkic languages, the ethnic name made of the structure of number+name is typical, e.g., Onogur ‘Ten-Ogurs’. According to the historian al-Mas^cüdi (10th century), the Turks called the Khazars Savirs.⁵³² Based on this data, in this Rovas inscription, the alternative name of the Khazars was ‘Six-Savirs’. According to the author of this book, the original ethnic name could have been the Six-Savirs if the ethnic name Khazar originated from the Roman title Caesar as Röna-Tas stated (Sect. 3.5.1).⁵³³ In the last word (from the right), the character betweenR /r/ andNG /ŋ/ is hard to read; Vekony reconstructed it asARCHED I /ï/.

The parsing of the inscription is heavy on nouns and light on morphology, resembling other unsuccessful reconstructions with spurious flavor. Typical Türkic phrase is heavy on conjugations and declensions, and light on nominals and verbs. In addition to being reasonable phonetically, a phrase must also make sense.

In Türkic don is “cold, icy”, Herodotus said: “icy Tanais”, hence here the inscription says in Savir language “icy, white”. Nobody knows what was the As-Alan name for Don, the inherited from the locals default As-Alan name is also the Türkic Don “cold, icy”. The conflation of the names As, a part of the known As-Tokhar confederation, and the umbrella term Alan “Steppe-man, Flatlander” is disturbing. Tokhars, for example, were Alans, but not the Ases.

From Herodotus times to end of the 10th c. AD the local name for Dnieper, and also for Buh, was a range of allophones, reflexes, and calques: from Βορυσθένης (Borusthenes), Borysthenes to Burichai, meaning “Bear stream”. The Buh was discriminated by adjective “White”: Κουβοΰ “White Winding”, where Kuu is “White”. Khazars were not natives of the Dnieper basin, it is inconceivable that they would come up with new names for old places of common knowledge. Ödüg in the inscription may be “selected, chosen” river, i.e. a fort on the suitable river, cf. ödür- “select, choose”. That is, if the parsing is accurate.

The Arabs called Savirs, and numerous other ethnically non-Khazar people, Khazars. Savirs (Suvars) are known in the Caucasus much earlier than the Khazars, and they were a ruling tribe before the advent of the Khazars. The Arabs, like the Khazars, were late comers, the Arab term Khazar is solely political and separate from the ethnic terms. The term Savir is ethnic and political, in addition to being specifically ethnic, it also covered numerous ethnicities in the Savir domain: Kayis, Masguts/Massagets/Alans, Huns, Bulgars, etc., and all local sedentary tribes.

The Six-Savirs confederation has no relation to the Khazars other than it has transferred to the Khazar Kaganate their suzerain authority after their capital Samandar was taken by the Arab commander Maslama in course of 727/728 campaign.

Whatever speculations on the origin of the name Khazar are, the names Khazar and Six-Savirs are not related other than by political correlation.

Second row: Ud Didü On /uð dⁱðy °n/ was a personal name. Another possible transcription of /dⁱðy/ is /dⁱdy/. This was one of the Khazar versions of the ethnic name Jewish. Its other version was found in the Alsöszentmihâlyfalva Inscription (Fig. 7.2.14-2). Both words originated from Hebrew.⁵³⁴

Here we have another disconnect: Suvars would unlikely use a Hebrew word for Jews when they have a Caucasian lingua franca word Djuga (cf. Djugashvilli “Son of Jew”) , known and understood by everybody in the Caucasus. The ud or uth is “cow”, on is “ten”, as a noun phrase the compound is unrealistic “Ten-cow Jew” or some perturbation of these three words as a name for a Jewish chief artisan in charge of decoration of the fortress. He also had a title, in a third-person reference naming a title is expected, naming a personal name is out of place.

530 Vekony 2004b
531 Golden 1979. p. 173; Golden 1980. p. 252 532 DeGoeje 1893 1894, p. 120
533 Röna-Tas 1999a, p. 228
163

Third row: The second character from the right needs reconstruction: it was N GH /ɣ/ orFORKED E /a/, and therefore, the first word was Onagh N /on^aɣ/ or Ona Y /ona/.

The inscription contains much information: the writer was Ineg, the builder was Ud Didü On, the nation was Six-Savirs, and the fortress was the Mansion of Onagh Tegin. The inscription declared the Khazars’ supremacy of the region between the Don and Dnieper rivers, the land of the As people. It is noteworthy that the Mayaki Fortress was near the Don River; this geographical fact supports the exactness of the transcription.

7.2.10. Mayak Fortress smaller inscriptions

In 1909, N. E. Makarenko found short Rovas inscriptions in the Mayaki Fortress. They are dated to the period from the late 8th to the 9th centuries.⁵³⁵ The drawings of the Khazar Rovas inscriptions were published by G.F. Turchaninov as well (Fig. 7.2.10-1,2,3,4,56).⁵³⁶ In the followings, the transcriptions are presented after each drawing.⁵³⁷

Figure 7.2.10-1: Drawing of the Khazar Rovas Inscription No. 1 of Mayak Fortress⁵³⁸

Table 7.2.10-1: Transcription of the Inscription No. I of Mayak Fortress

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	daðqajbin β... or daðqajbin k...
Translation from Common Turkic (Khazar)	‘I will write w ...’ or ‘I will write k ...’

The left hand character can beOPEN V /(β/b-/ orFORKED K /k/ as well. This inscription shows typical features of the Khazar language: in the beginning of the word the changes /j/ > /d/ and /z/ > /ð/ happened. However, the change /j/ > /d/ did not appear in every Khazar dialect; as it can be compared in the two versions of the name ‘Jewish’: /dⁱðy/ in the Mayak Fortress Relic (Table 7.2.9-1) and /jyedi/ in the Alsöszentmihâlyfalva Relic (Table 7.2.14-1). (Yedi, like Djuga, stands for Iudea. It is a common designation in Common Türkic outside of the Caucasus)

534 Vekony 2004a, pp. 221-222
535 Makarenko 1911, pp. 21-22, Fig. 19-22
536 Turchaninov 1964, pp. 75-76
537 Vekony 2004b
538 Vekony 2004b
539 Vekony 2004b
164

Figure 7.2.10-2: Drawing of the Khazar Rovas Inscription No. 2 of Mayak Fortress⁵³⁹

Table 7.2. JO-2: Transcription of the Inscription No. 2 of Mayak Fortress

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Ogur) transcription (left-to-right)	bethisetrim
Translation from Ogur	‘N. [member of a clan], I had it carved.’

The first two symbols of the inscription make up a tamga. Similarly, complex tamga symbols are known in other Rovas inscriptions, e.g. Needle Case of Jânoshida (Sec. 6.2.3).⁵⁴⁰ The glyph of the KRARCHED I /i/ is reversed in the inscription; however, mirroring glyphs is usual in KR.

There is a Common Turkic verb /beðiz/ meaning ‘carve’. The Ogur does not have the consonant /z/; hence it substituted /z/ with /s/. Therefore, the inscription is in Ogur language and the word /beðiz/ was a Common Turkic loanword. The word /beðizet/ is a causative form; its meaning: to have it carved. The /beðizetrim/ means ‘I had it carved’. Since the tamga means a clan-name, the meaning of the inscription is ‘N. [member of a clan], I had it carved’.

Tamga is a community property, and does not change as long as the clan exists, and long after its dispersion. Tamga

must be known from the Caucasus tamga catalogs, or else it is not an attested tamga, and could be a part of the inscription.

Figure 7.2.10-3: Drawing of the Khazar Rovas Inscription No. 3 of Mayak Fortress⁵⁴¹

Table 7.2.10-3: Transcription of the Inscription No. 3 of Mayak Fortress

Written with normalized Khazar Rovas font
IPA phonetic transcription	/nⁱband/
Common Turkic (As-Alan) transcription (left-to-right)	niband
Translation from As-Alan	‘Bonding’

The character is the ligature ofRAISED B /b/, Y A /a/ andN /n/. A special feature of this relic is that the inscription is on two stones, in a way that the half of the characterARCHED D /d/ is on one stone and its other half is on another stone fitting the first one. The purpose of the text is obvious: it is a building order showing where the bonding has to be made.

Figure 7.2.10-4: Drawing of the Rovas Inscription No. 4 of Mayak Fortress⁵⁴²

The character is the ligature of D CLOSE J /j/ and the mirrored glyph ofANGLED I /i/. In this inscription, the glyph , a mirrored variant of theOPEN M /m/ was in use. This relic is related to the text of the Table 7.2.10-3.

⁵⁴⁰    Vasary 1972, Fig. 1
⁵⁴¹    Makarenko, 1911, p. 28, Fig. M3
⁵⁴²    Vekony 2004b
165

Table 7.2.10-4: Transcription of the Inscription No. 4 of Mayak Fortress

Written with normalized Khazar Rovas font
IPA phonetic transcription	/s^am^ajik^am/
Common Turkic (Kypchak) transcription (left-to-right)	samajikam
Translation from As or Alan	‘Let us assemble it.’

Figure 7.2.10-5a: The photograph of the Khazar Rovas Inscription No. 5 of Mayak Fortress made by S. Ya. Baichorov, and b) its drawing⁵⁴³

a)	b)

In this inscription, the glyph a mirrored variant of theARCHED I /i/ï/ was used.

Table 7.2.10-5: Transcription of the Inscription No. 5 of Mayak Fortress

The /-aq/ is a Turkic diminutive suffix of the personal name /on/, it means ‘ten’.⁵⁴⁴

Written with normalized Khazar Rovas font
IPA phonetic transcription	/onaq ^ɛbi/
Common Turkic (Khazar) transcription (left-to-right)	onaq ebi
Translation from Common Turkic (Khazar)	‘Mansion of Onaq’

7.2.11. Bilingual Khazar Coin of Sweden

An inscribed coin was found in Sweden with Arabic and Khazar Runiform characters (Fig. 7.2.11-la, b, and c)⁵⁴⁵

Figure 7.2.11-1 a, b) The Arabic and c) Khazar Runiform inscriptions on the Bilingual Khazar Coin of Sweden

a)	b)	c)

The Arabic inscription in Fig. 1.2.11-la means blessing or wealth,⁵⁴⁶ and in Fig. 7.2.11-lb a personal name: /nâr yık/. The transcription of the Khazar inscription of Fig. 7.2.11-1c is presented in Table 7.2.11-1.

543    Baichorov 1989, Table 120
544    Gombocz 1915, pp. 38-39
545    Vekony 1997a
546    Charaf 2010
166

Table 7.2.11-1: Transcription of the Khazar Runiform inscriptions on the Bilingual Khazar Coin of Sweden

Written with normalized Khazar Rovas font	First row: ; second row:
IPA phonetic transcription	First row: /^øβy/; second row: /ⁱk øðⁱm/
Common Turkic (Kypchak) transcription (left-to-right)	aishermith bith ishermith boshadin
Translation from Common Turkic (Khazar)	First row: 'Extolling’; second row: 'Ik, myself’

The word /ik/ was a personal pronoun. Vekony claimed that this word is in Common Turkic (surely Khazar); since its Ogur version would be /yik/.⁵⁴⁷ However, Gombocz stated that the /yik/ in the Arabic text corresponded to the Common Turkic (Khazar) /ik/.⁵⁴⁸

The runic Khazar

/ik/ Ik “I” is an allophone of the German Ich, Ih “I”, and English I “I”. The other Germanic, Celtic, Latin, Greek forms are all allophones of the Khazar Ik: Anglo-Saxon ik (spelled ic), Dan., Norse jeg, Dan. ik, Icl. eg; Ir. agam, me ag “I mine, me I”, Welsh wyf, Lat. ego, Gk ego (εγώ), ekho (έχω), all “I”. The Irish form is peculiar, it literally combines semantics of “inner” with the semantics of “me”. The form ič of the Türkic word “inner” is the base for the Türkic ič “I” and all its known allophones across Eurasia. The me (min, men, bin, ben) is the objective form of “I”, and probably it was in circulation long before the form ič has appeared. The Irish has preserved frozen archaic forms from the time when ič “I” without definition “me”, “mine” still denoted “inner”

The /nar/ was the name of a Kagan; see Kagan Nar /n^ar x^aɣ^an/ the Ogur inscription of the Mayaki Amphora (Fig. 7.2.2-2). The Arabic inscription of the coin contains the full name of the Kagan: Nar Ik in Common Turkic (Khazar). (I.e. “I (myself) Nar”) To summarize, the Arabic and the Khazar Runiform inscriptions on the coin are essentially identical.

7.2.12. Cattle skull from the kurgan Ermen Tolga

In 1970, a fragmented cattle skull was found in the plundered central burial of the kurgan Ermen Tolga, 15 km from Elista (Capital of Kalmykia, Russia).⁵⁴⁹ In the scientific literature, it is several times called bull skull. The relic is from the period between the 8th and 10th centuries. There are inscriptions on 11 of the 16 fragments of the skull. According to S.G. Klyashtorny and I. Vâsâry, there are 80 symbols on the fragments.⁵⁵⁰ Fig. 7.2.12-1 presents the third row of the three-row inscription; Fig.7.2.12-2 shows the drawing of the reconstructed cattle skull; and in Fig. 7.2.12-3 there is the drawing of the reconstructed skull with the inscriptions. The cattle skull was reconstructed by the archaeologist I. Erdelyi and the archaeozoologist I. Takâcs.⁵⁵¹ According to Takâcs, the age of the cattle was adultis/maturus.⁵⁵² According to Erdelyi, the inscription was written by two persons sitting opposite to each other, since the style of two rows of the inscription significantly differs from the third row being upside down.⁵⁵³ Table 7.2.12-1 presents the transcription of the texts.⁵⁵⁴

Figure 7.2.12-1: The fragments of the cattle skull of Ermen Tolga.⁵⁵⁵On this relic, the glyph - a mirrored variant of theOPEN M /m/ and the glyph - a variant of theARCHED UE /ø/y/ were used.

547    Vekony 2004a, p. 282
548    Gombocz 1915, p. 48; Gombocz 1908, p. 22; Gombocz 1912
549    Erdniev 1982, Fig. 2-7; Kyzlasov 1990a, Fig. 13. 14. pp. 42-47
550    Klyashtomy & Vâsâry 1987
551    Erdelyi 1991, pp. 150-153
552    Erdelyi 1991, p. 150
553    Erdelyi 2010
554   Vekony 2004b
167

Figure 7.2.12-2: Reconstruction of the cattle skull of Ermen Tolga	Figure 7.2.12-3: The drawing of the fragmented Inscription of Ermen Tolga

Table 7.2.12-1: Transcription of the fragmented Inscription of Ermen Tolga

Written with normalized KR font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	Second row: /deb idi deb ö abam yge eba ö.../ First row: /ud byshi ab[a].../ Third row: /...a ö : öme bash de abam byshi ish : maŋa edith ... ö .../
Translation from Common Turkic (Khazar)	Second row: ‘Lord spirit, spirit, think! My father, wise house-father, think!’ First row: ’ . bovine-wizard father ...’ Third row: ’ Oh : think! Don’t think, Head! Eat! My wizard father, drink! For my highly ... think ...’

555 Vekony 1997a, cover inside
168

The word /deβ/ ‘spirit’ was an As or Alan loanword in the Khazar language (< Old Persian /daiva/). The expression /^ɛb-^aba/ ‘house-father’ meant the spirit of the house, its defender. The inscription recorded a magic text, being probably rhythmically chanted during a shaman ritual.⁵⁵⁶

With the polysemantics innate to the Türkic lexus, even a correct reading has a spectrum of interpretations. To add to the unknown some unverifiable parsing, unattested borrowings, and primitive religious biases really hurts the credence of the interpretation. Kurgans are the tools supplied to a deceased relative to reach the Almighty Tengri for reincarnation, the kams (“shamans” in Vekony's lingo) with their black magic are not admitted to the Tengrian rituals. There is no more sacred ritual than the send-off to the Almighty, and a wealth of dedicatory grave inscriptions give examples of what can be expected at a kurgan grave. The Ermen Tolga grave is located in the territory of the Khazar domain, in the land of Saklans, which is connected with the “As or Alan” only in the mythopoeic thought of some creative patriotic scientists. The very idea that Khazars had an Old Persian loanword borrowed from the “As or Alan” language attests to the surreal mythopoeic thought underlying interpretations.

7.2.13. Quiver Inscription of Homokmegy-Halom in Turkic

The Khazar Rovas inscription on a bone-plate covering a quiver was discovered by the archaeologist I. Dienes (Fig. 7.2.13-1), and was researched by Vasil’ev among others.⁵⁵⁷ The relic is from grave No. 6 of a cemetery dating to the period of the Magyars’ Landtaking (10th century).⁵⁵⁸ This cemetery is in Homokmegy-Halom (Hungary, see Map 6.2-1), one of the centers of the Magyars led by the Grand Prince Ârpâd. According to I. Dienes, this type of quiver was unknown in the Avar Age; therefore, it surely was the product of the Magyars, and the owner of the quiver was a member of the Magyar Army.⁵⁵⁹ Fig. 7.2.13-2 presents the drawing of its inscription.⁵⁶⁰

Figure 7.2.13-1: The bone-plate of the Quiver of Homokmegy-Halom (10^lh century)
Positive image	Negative image

Figure 7.2.13-2: The drawing of the Inscription of Homokmegy-Halom⁵⁶¹

The word /s^aɣd^aɣ/ ’quiver’ is a loanword in Turkic; it originated from the Iranian /sa:xtag/ ‘garniture, equipment’. The transcription in Table 7.2.13-1 was made by G. Vekony and it was improved by I. Vâsâry.⁵⁶² According to the author, the KR CLOSE G /ɣ/ surely originated from the Inscriptional Pahlavi QOPH /q/. Presumably, the Iranian word /sa:xtag/ was accepted by Turkic with the /x/ > /q/ substitution in the middle of the word. That could be the reason why the KR CLOSE G originated from the Inscriptional Pahlavi QOPH /q/ was applied; and probably, the CLOSE G represented /q/ at that time. Later, the /q/ > /ɣ/ transition happened. Due to the conservativity of the writing, the CLOSE G remained in this word, but it already represented /ɣ/.

Table 7.2.13-1: Transcription of the Inscription of Homokmegy-Halom

Written with normalized Khazar Rovas font
IPA phonetic transcription	/^on^oglⁱɣ s^aɣd^aɣⁱn b^as /
Common Turkic (Bulgar?) transcription (left-to-right)	/onogliɣ saɣdaɣin bas/
Translation from Turkic	‘Triumph with [a] quiver of ten arrows’

Ligeti L. Mots de civilisation de Haute Asie en transcription chinoise // Acta Orientalia. 1950. pp. 141-149, discussed the Hunnic word sagdag meaning boots, given in Chinese transcription, and speculated on its origin. It turned out that that word was borrowed from the Türkic, likely Kazan Tatars, into Russian, in the form sagaidak (сагайдак), and was widely used in Russia in 16th-17th cc [L. N. Gumilev, 1960, History of Hunnu people, Moscow, "Science", Ch.3]. Sagdag refers to the bootleg of the boots, where the mounted archers kept their arrows in addition to the regular quiver, and the inscription refers to the ten arrows in the bootleg of the boot. E. Pulleybank speculated that that word may be a Ket word, but that idea was discarded based on the ethnological specifics of the hunter-gatherer practice. The Hunnic word could have been borrowed into Old Persian from the Scythians and Kangars in the Persian service, but at any case Bulgars had no need to borrow foreign lexicon for tools innate to their nomadic life. The scholarly speculations are an exercise in fatuity, other than that the Old Persian and Türkic scripts were adopted form the same source, albeit at different times, since the Persians switched from the cuneiform to alphabetical writing relatively late.

According to Omniglot http://www.omniglot.com/writing/mpersian.htm, Inscriptional Pahlavi did not have a character q:

The polysemy of the word bas- is a good example of possible readings and misreadings. The bas- is “to press”, “to stamp (with a stamp)”, “suppress, overcome”, “attack”, “take over”. Probably, in this case the semantics of “suppress, overcome” was selected for “triumph”. The possessive suffix -in points in a right direction: “My 10-arrow sagdag (would) prevail (over enemy)”.

556   Vekony 2004b
557    Vasil'ev 1983
558   Dienes 1972. p. 67; Dienes 1992, p. 31
559    Dienes 1992, p. 39
560    Dienes 1992. p. 31-40.
561    Vekony 1992a, p. 42
562    Vekony 2004a, pp. 56-47; Vâsâry 2010-2011
169

According to Vâsâry, the language of the inscription could be either Common Turkic or Ogur.⁵⁶³ This inscription proves that Turkic was used as well by the army of the Grand Prince Arpad. This supports the fact that the culture of the landtaking Magyars had a strong Turkic heritage. For instance, the mythological bird turul of the Ârpâd Dynasty originated from the Turkic togril, which was a bird name.⁵⁶⁴ Another fact is that between 948 and 950, the Magyar leader Zombor visited Byzantium, and he returned with the missionary bishop Hierotheos, whose title was ‘Bishop of Turkia'.⁵⁶⁵

Each Türkic tribe had its ongon, a tribal bird, that distinguished the tribe from other tribes. Ongon has a great diagnostic value, knowing the ongon can limit the number of candidates to very small number.

7.2.14. Stone Inscription of Alsöszentmihâlyfalva in Khazar

The inscription on a building stone was found in Alsöszentmihâlyfalva (present-day Mihai Viteazul, Transylvania, Romania, see Map 6.2-l)\ its photograph was made by E. Benkö (Fig. 7.2.14-1).⁵⁶⁶ The stone was an ancient Roman building stone - proved by the leaf-symbol, a frequently applied ornamental element of ancient Roman inscriptions - reused in the 10th century. Alsöszentmihâlyfalva was located on the territory where the late Province Dacia existed up to the middle of the 3^rd century. Denes showed that the Khavars (Kubars, Kuvars, Kovars, Cuvers) (Khazar rebels joined the Hungarians in the 9^lh century⁵⁶⁷) probably settled in this region (at that time Transylvania).⁵⁶⁸ In some parts of Hungary, there is data of the Khavars even from the 13^lh century.⁵⁶⁹ The drawing of the Khazar Rovas inscription and its transcription are presented in Fig. 7.2.14-2 and Table 7.2.14-1.⁵⁷⁰

Figure 7.2.14-1: The Alsöszentmihâlyfalva Inscription (920-952)
Positive image	Negative image

563   Vâsâry 2010-2011
564   Vâsâry 2003
565   Tempfli 2000
566   Benkö, E„ 1972a, p. 453 & Appendix; 1972b
567   Kristö & Makk 2001, p. 52
568   Denes 1984-1985, p. 573
569   Györffy 1990, p. 50
570 Moravcsik 1984, p. 85; Vekony 1987a, pp. 108-117; 1997, pp. 108-117; 2004a, pp. 217-230
170

Figure 7.2.14-2: Drawing of the Alsöszentmihâlyfalva Inscription (920-952).

Table 7.2.14-1: Transcription of the Alsöszentmihâlyfalva Inscription

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Khazar) transcription (left-to-right)	First row: /^ebi ^atlⁱɣ/ Second row: /jyedi kyr qar^aj/
Translation from Common Turkic (Khazar)	First row: ‘His mansion is famous’ Second row: ‘Jüedi Kür (the) Karaite’

The word /eb/ was used for larger buildings in the Khazar Rovas building inscriptions;⁵⁷¹ therefore, its meaning is ‘mansion’. The building inscriptions are typical in KR relics (e.g., Sec. 7.2.9 and 10). The word /jy^edj/ meant Jewish in the Khazar language, in this case it was probably used as a personal name. The /kyr/ means ‘courageous’ and ‘fearless’ in Turkic,⁵⁷² such personal names are typical. Therefore, the second row or at least the first two words (Jüedi Kür) formed the name of a Khavar leader. Presumably, the word Karaite was also a personal name. The symbol is a typical punctuation mark in KR and the symbol in the third row was probably a tamga, according to Vekony.

The Karaite Judaism is a branch of the Jewish religion accepting the five books of Moses but not the Talmud, oppositely to the Rabbinic Judaism. Between the 10th and 11th centuries, a large number of Karaite works were produced in the eastern part of the Muslim world. Presumably, there were a significant number of Karaite people in the Khazar Kaganate as well.⁵⁷³ It is noteworthy that Jüedi Kür emphasized that he was Karaite, in his language: /qar^aj/.

According to Vekony, the inscription was made between 920 and 952 based on the following facts. At the beginning of the 10th century, the Khavars (Kubars, Kuvars, Kovars, Cuvers) still served as guards at the western borders of Hungary. Therefore, they settled down in Transylvania only later.⁵⁷⁴ In 952, the local leader of Transylvania titled Gyula and his followers (among them the Khavars) converted to Christianity. Consequently, this Rovas drawing was obviously made before that time.⁵⁷⁵

To understand the inscription, a little primer on Kubars and other Türkic tribes is needed, see Convergence - Türkic folks and Hungarians. Kubars. In a nutshell, Kubars is a distorted term for the autonomy of Cuver, a fourth son of Kurbat. The name Kubars describes the Cuver's tribes, the one large Finno-Ugrian and six to eight smaller Türkic tribes. Somehow, his principality had survived for 200 years or 8 generations, from the death of Kurbat (ca 660s) to evacuation from Levedia (ca 850-860). Geographically, Levedia is synonymous with Kara (Western, or Black) Bulgar. No Khazar tribes migrated with Kubars, so the attribution of the inscription to the Khazars is a misnomer. The tribes that may be candidates for the inscription are Seklers and Kubars, the last largely composed of Bulgars, Suvars with their princely tribe Baryn, and Ases. The administrative division of the Kubars may be ignored, although each administrative unit probably had an ethnic majority, these units are known as Baryn (Suvar), As, Örs (Ugro-Finns), Tarkhan, Varsan (Avars), Sabir (Suvar), Kaliz-Kalaz (Alats, Khalaj), Khazar, Sekler. Seklers and Bulgars would be most likely candidates if not for the testimony of the inscription itself.

Two peculiarities of the inscription attract attention. The symbol , if it is a word separator, is located at the end of the inscription, where there are no words to separate. Could the interpretation of this grapheme as a word separator be erroneous?

The grapheme in the third line, interpreted as a tamga, appears to have a colon word-separator in front of it, i.e. it may be a text. As a tamga, it is a double forward slash, or a thunderbolt. The double forward slash belongs to Oguz tribes Karkyn and Kosyp, Kirgiz tribes Kosh and Kerki. The thunderbolt belongs to Oguz tribe Igdır, Kazakh Kangars. It is not a Y-looking Karaite tamga. Could a Kipchak dignitary be a leader of Karaites? It appears that as tamga it may be inconsistent with the text of the inscription.

The words Kür Karaite appear to be a title, “Jewish Karaite Prince”. The title Kür comes in a variety of dialectal allophones, from Kir to Chur. The princely inscription would come with his tamga quite naturally, the absence of tamga would be oddball. Personal names are a rarity in the Türkic milieu. The Türkic naming tradition leads a person through a chain of names, and they can change on random occasions. Title-names are used at maturity. The scholars who came up with the idea of personal names suggest to apply their notions of the names to a subject outside of their notions.

The title Gyula of the local leader of Transylvania is a peculiar form of Ulug, like Ulug Khan or Ulubek, it stands for Prime Minister (of Transylvania). The initial g- is the Ogur form of the title, Gulug, known from Hunnic titles in Chinese transcriptions.

571 Vekony 1997a, p. 110
572 DTS, p. 328
573 Karaism: Origin & History (Web page)
574 Vekony 2004a, p. 228
575 Moravcsik 1984, p. 85
576 Golb& Pristak 1982
171

7.2.15. The Khazar Rovas inscription on the Kievan Letter

In 1962, N. Golb discovered a non-Hebrew inscription on the bottom left corner of the Kievan Letter.⁵⁷⁶ The authenticity of the letter was examined by Röna-Tas.⁵⁷⁷ The photocopy of the Khazar Rovas inscription on the Kievan Letter is presented in Fig. 7.2.15-1. According to O. Pritsak, the Kievan Letter was made before 930 AD.⁵⁷⁸ Iranist J. Harmatta proved that the Kievan Letter was not written in Kiev.⁵⁷⁹ In fact, it was addressed to Kiev.⁵⁸⁰ There is this sentence in it: “we (the singers of the letter) inform you, the community of Kiev”.⁵⁸¹ The place of writing the Hebrew text was named in the letter *K’.MN, which can be reconstructed as Karman.⁵⁸² Albeit the Khazar-Jewish letter writing orthography used the denomination madinah ‘city’ for the Khazar capital, it probably corresponds to Karman, according to Vekony. In other words, this name was probably used to specify the Khazar capital and the letter was written there.

First O. Pritsak gave a transcription of the non-Hebrew inscription,⁵⁸³ which was refined by L. Ligeti;⁵⁸⁴ and this transcription was argued by A. Röna-Tas.⁵⁸⁵ According to him, the Kievan Letter is from 955 to 961 due to historical reasons.⁵⁸⁶ Fig. 7.2.15-2 shows the precise drawing copied directly from the photograph of the original text.⁵⁸⁷

Figure 7.2.15-1: Enlargement of the original Khazar Rovas inscription on the Kievan Letter⁵⁸⁸
Positive image	Negative image

Figure 7.2.15-2: Drawing of the Khazar Rovas inscription on the Kievan Letter⁵⁸⁹

Table 7.2.15-1: Transcription of the Khazar Rovas inscription on the Kievan Letter

Written with Khazar Rovas font
Written with normalized Khazar Rovas font without ligatures
IPA phonetic transcription	/oɣ^ïðïq ilⁱk/
Common Turkic (Khazar) transcription (left-to-right)	/oɣïthïq ilik/
Translation from Common Turkic (Khazar)	‘We have read. Ilik.’

Ilik (Ulug, Ulug-Bek) was the name of the vice-king (Prime Minister) in the Khazar Kaganate (Sect. 7.2.3). The vice-king (Prime Minister) was in charge of the day-to-day operations.⁵⁹⁰ Generally, the version yilik is known for the vice-king (Prime Minister)⁵⁹¹, but Z. Gombocz proved that the Arabic (phonetical) form yilik ‘vice-king’ can also be transcribed as ilik.⁵⁹²

577 Röna-Tas 1982a, p. 350
578 Golb & Pristak 1982, 42-3, p. 71
579 Harmatta 1997, pp. 119-140
580 Vekony 2004a, p. 277
581 Golb & Pritsak 1982, p. 6
582 Vekony 2004a, p. 277
583 Golb & Pristak 1982
584 Ligeti 1984, pp. 10-17
585 Röna-Tas 1983-84
586 Röna-Tas 2001, p. 181
587 Vekony 2004a, p. 278
588 Wikimedia Commons, web site: Okurttm-Khazar.png. Public Domain. Retrieved in 2011 from http://commons.wikimedia.org/wiki/File:Okurum-Khazar.png. A part of the image.
589 Vekony 2004a, p. 278
590 Golden 1984, pp. 474-486
591 Golden 1980. p. 162, pp. 184-185
592 Gombocz 1915, p. 48; Gombocz 1908, p. 22; Gombocz 1912
172

Other versions of ilik were ilek or ilig.⁵⁹³ The etymology of the ilig is il + lig, its meaning is ‘who has people’ (rather lit.“national leader”, il- “country, state, nation”, -lig/-lïq/-lik/-luq/-lük denoun suffix to form abstract and purposeful nouns), and this dignitary name was typical in the Turkic nations. The title was also used as a personal name. According to Constantine VII Porphyrogenitus, Ielekh (version of Ilik) was the personal name of one of the sons of the Grand Prince Ârpâd. The name Ielekh survived in the geographical name as Üllö Village near Budapest (Hungary).⁵⁹⁴ In the Gesta Hungarorum, the first record of the early Hungarian history, the name Hulec /hylek/ was mentioned as member of the Ârpâd-dynasty. The name Hulec is the labialized variant of the Turkic name /ilik/,⁵⁹⁵ as Vekony also stated.⁵⁹⁶

In a dialectal speech, Ulu/Ulug/Uluɣ and Ilik/Ilek/Ilig can be homophonous and conflated, since they designate the same position. Given that common terminology across Kaganate tended to level out, the word Ilig for “national leader” is well substantiated, but the ubiquitous “Bek” for the same position is a short for Ulug Bek “Great Prince”, which leaves both interpretations viable.

The Khazar Rovas inscription /oɣ^ïðïq ilⁱk/ ‘We have read. Ilik’ in Fig. 7.2.15-1 most probably functioned as reconfirmation of the letter. The verb /oɣ^ïðïq/ ‘we have read’ demonstrates the result of the /q/ > /ɣ/ transition specific for Kypchak. As the inscription preserved the official Khazar language, it proves that it belonged to the Kypchak (branch of Common Turkic language). It is noteworthy that Erdal also supposed that Khazar was Common Turkic and not Ogur.⁵⁹⁷

“The /q/ > /ɣ/ transition specific for Kypchak” indirectly asserts that Kypchak languages were casual in the Kaganate long before the arrival of Kipchaks in the last years of the first millennium AD, i.e. that the leading elite of the Kaganate, consisting of the Bulgars and Suvars, were speaking Kypchak-type languages. This contradicts the research that stipulates, first, that the Bulgar, Suvar, and Khazar languages were about the same, and second, that they belonged to the Ogur branch.

Kipchaks followed Oguzes into the Eastern Europe, Oguzes ca 750, and Kipchaks ca 1000. Oguzes could bring over the Oguz/Kipchak-type language, say, at about 800, but to override the ingrained Ogur speech and become a lingua franca reaching the upper echelons of the power in the Kaganate is a very long shot. Both Oguzes and Kipchaks remained separate cohesive communities long after the Kaganate dissipated. That attests that neither Oguzes nor Kipchaks could penetrate the hereditary ruling elite of the Kaganate and force their language on the Kaganate's rulers.

The Khazar “/q/ > /ɣ/ transition specific for Kypchak” needs to fit into the linguo-historical picture of the times.

7.2.16. Pottery of Bilyar

In 1985, Kochkina published the inscription on the handle of a vessel (Fig. 7.2.16-1), estimating its origin to the 11th—12th centuries.⁵⁹⁸ Khalikov dated the relic to the 12th—13th centuries.⁵⁹⁹ The leftmost character of the inscription is surely the Khazar Rovas punctuation mark : Colon. A possible partial transcription is presented in Table 7.2.16-1. The reading /d^yʃ/ ‘dream’ was proposed by the author. Turkologist I. Vâsâry improved it to /dyɬ/ and gave another possible meaning: ‘noon’⁶⁰⁰ He stated that the /dyɬ/ could be the Ogur version of the Common Turkic /d^yʃ/ ‘dream/noon’. The second part of the inscription is still unsolved. The fourth character of the inscription from the right is probably theCLOSE T /t/; however, evenCLOSE R /r/ might be supposed.

The language of the inscription is most probably Ogur, since Bilyar was in the middle of Volga Bulgaria. The sound value of the KRSHL /ɬ/ indicates the transitional sound between the Common Turkic /ʃ/ and Chuvash /l/. Presumably, the voiceless alveolar lateral fricative /ɬ/ was in use in Ogur.⁶⁰¹ The /ɬ/ exists even today in the Finno-Ugric Khanty language as well. Note that the dental version of this sound is denoted also with /ɬ/ in IPA. The dental version exists today in the Celtic Welsh; it is Latinized as ll.⁶⁰² Moreover, the sound /ɬ/ exists in the Caucasian Avar as well.

593 Khazarian Names, web site, www.klwzaria.com
594 Vâsâry 2003
595 Ligeti 1979, p. 261
596 Vekony 2004a. p. 282
597 Erdal 1999 in Golden, Ben-Shammai & Röna-Tas 1999, pp. 76-107
598 Kochkina 1983; 1985 pp. 54-80
599 Khalikov 1988
600 Vâsâry 2010-2011
601 Vâsâry 2010-2011
602 Vâsâry 2010-2011
173

Figure 7.2.16-1: The Bilyar Inscription (11th—13th centuries)
Positive image	Negative image

Table 7.2.16-1: The partial transcription of the Bilyar Relic βŋïøöüɣðʃəɛ

Written with normalized Khazar Rovas font
IPA phonetic transcription
Common Turkic (Ogur) transcription (left-to-right)	/dyl (dysh)/ /b...t...n/ (/b...r...n/)
Translation from Ogur	‘dream/noon’ (second part is undeciphered)

It appears that the best match for a jar would be something connected with liquids and consumption. The surviving part of inscription is its tail end, a punch line of the whole phrase. The phrase was probably a rhymed couplet. In that line of thinking toş-/doş-/tuş-/duş- “to fill” for the dysh would be quite suitable semantically and phonetically. The last word in the phrase most likely, but not obligatory, is a verb. The last word b...t...n (b...r...n) may have plenty of possibilities: bat- “to bathe”, bit-/büt- “starting, setting off”, “heal”, bat “hurriedly”, bod “body, stature”, “color”, but “thigh, leg”, “treasure”, böd “throne”, bedü- “grow”, bödi:- “to dance”, bar- “go, leave”, ber- “give”, bur- “to smell, to steam”, bere: “length”, berü “here”, bedze “adorn”, and quite a few more. The jar itself is some embodiment of /b/p...t/ “pot”. That choice gives plenty of possibilities to the artisan poet to coin a ringing phrase that attracts buyers. For example, “XXX... fill the pot”, “XXX... fill (to the rim) to give (present, for the road, etc)”, “XXX... fill to grow” “XXX... fill to the (full) length”, etc.

Notably, the notion of “noon” is a metaphorical from a metaphor, to “stop” > “stop to rest” > “stop for rest at noon” > “noon”, its essence is not noon, but stop to have a rest, it is a long and winding way to find a phonetical match conditioned on a supposition that the /sh/ can be altered to /ll/. From the above examples it is clear that in addition to “noon”, such emendation opens a whole slate of new possibilities, all of them unprovable

174