EDWARD TRYJARSKI

(Warsaw)

The Unity or Multiplicity of Runic Scripts. An Account of the

Attempt to Isolate a South Yenisei Alphabet

Present-day research on the general character, origin and chronology of

the runic system of writing in Asia and southeastern Europe has been con-
fronted with certain recurring questions that cannot be left unanswered. One
of them is the following: are we dealing with the development of one system
which originated in a given location and evolved throughout the ages or with
that of several more or less similar alphabets which, although cognate in one
way or another, have sufficiently long histories to be treated as independent
units?
The notion that the Turkic runic alphabet is homogeneous took shape
during the last two centuries and became a kind of dogma, especially when
great discoveries were being made in Mongolia and in East Turkestan, during
the first attempts at deciphering the unknown script. Some less impressive
and shorter inscriptions but for scholars almost as important were discovered
in the Yenisei Basin by D. G. Messerschmidt's expedition in 1721-1722.
Notwithstanding the existence of a number of different signs, there was no
serious doubt about the presumption that the scripts from Mongolia, the
Yenisei Basin and the Talas area basically formed one alphabetical system
with two or three variants. The notion that they are genetically related
seemed evident to the majority of scholars. Soon after a new group of texts
had been revealed in Semirechie, W. Radloff emphasized the existence of
two alphabets: the Orkhon alphabet and the Yenisei alphabet. 1 After some
time, S. E. Malov declared his support for the existence of the Talas alphabet.
Then as now the term Orkhon-Yenisei alphabet was commonly used to de-
note this whole system of writing. Consequently, the newly revealed different
signs, the number of which had increased as a result of further expeditions,
were regarded as local variants or as individual innovations of the inscrip-
tions' scribes.

1
Cf. I. L. Kyzlasov, Runiceskie pis'mennosti evrazijskikh siepej. Moskva, 1994:8.
366 Edward Tryjarski

This theoretical attitude had important practical consequences since, in

the course of time, the phonetic values of the classical signs from Orkhon,
Ongin and Yenisei were for no real reason applied to certain, seemingly
identical, characters of the inscriptions later discovered in central Asia and
Europe. This method of deciphering did not always bring satisfactory results.
The problem of deciphering the texts gave rise to certain other important
questions, one of them being: Is Orkhon script older than that of Yenisei, or
vice versa? The question was debated for a century, mainly in favour of
Yenisei script which, having more irregular shapes in its characters, was
considered to be more ancient. Therefore, it was supposed that Yenisei script
had undergone a deep transformation and become Orkhon script. Only in
1960 did L. R. Kyzlasov declare that, unlike W. Radloff, P. M. Melioranskij
and S. E. Malov, he believed Yenisei script to be later than Orkhon script. 2
More recently, A. Rona-Tas has made an attempt to be more precise: "[...]
the Yenisei inscriptions are simultaneous with or later than those of Mon-
golia", he writes. 3 Several generations of Turkologists believed that Orkhon-
Yenisei script had developed from Iranian, which was of Semitic and
probably of Aramaic origin and that its variants were substantially due to the
material or instruments used for writing. In particular, cursive script was
considered to be the result of writing on the parchment with a brush. How-
ever, no Turkologists were able to explain the existence of double signs used
to denote the same phonemes. A simple examination of the list of characters
indicates the existence of different signs for t1, q, m, nt and some others, a
phenomenon that could not be explained by either the alphabetic or syllabic
nature of this script. The situation was especially unclear due to several
specific complementary signs revealed in some new inscriptions originating
mainly from Kirghizstan and Europe. We need only remember an embarrass-
ing "Pecheneg ladder" that was detected on the wooden stick from Acyq-Tas.
S. E. Malov much hesitated to determine that inscription as belonging to the

L. R. Kyzlasov, "Novaja datirovka pamjatnikov enisejskoj pis'mennosti" Sovetskaja

Arkheologija, 4 (1960), 3:93-120; the same, "O datirovke pamjatnikov enisejskoj
pis'mennosti" Sovetskaja Arkheologija, 9 (1965), 3:38-49; the same, "Kogda izcezla
enisejskaja runiceskaja pis'mennost' juznosibirskikh tjurkov" Vestnik Moskovskogo
Universiteta, ser. 8, Istorija, 1992, 6:29-36. Cf. G. Clauson, "The Origin of the Turkish
"Runic" Alphabet" AO XXXII, 1970:53.
3
A. Rona-Tas, An Introduction to Turkology, Szeged, 1991:56-57.
 The Unity or Multiplicity of Runic Scripts 367

Turkic culture. 4 Also A. M. Scerbak was of the opinion that the signs of the
inscription do not belong to the Turkic alphabet. S. G. Kljastornyj suggested,
in 1987, the existence of two systems of runic script: a central Asiatic and an
eastern European. 5 This notwithstanding the old practical terms: "European
runes" and "Asiatic runes" are still in general use.
In the meantime, the problem of the runic alphabet of the European zone
became increasingly complicated. In their edition of short inscriptions on the
bull skull from Elista, S. G. Kljastornyj and I. Vásáry wrote in 1987:

"The inscription was written with a variant of the East-European runic script (in
the following EER). The area where this EER was in use comprises the steppe zone
of South-East Europe between the Volga (Lower and Middle Volga Region) and the
Danube Basin (territories of present-day Roumania, Bulgaria and Hungary)."6

In another passage, the same authors attempted to organize the new

epigraphic material and our knowledge of it:

"The first documents of the EER that has become known to the scholarly world,
are the inscriptions [...] of Nagy-Szent-Miklós found in 1799. [...] Recently, a new
inscription has been discovered on the amphora from Majaki, where the alphabet is
identical with that of Nagy-Szent-Miklós. Another type of the EER is represented by
what is known as the Khazar script. The documents of this script have become
known from the 1930s. This type of the EER, though evidently similar to the Nagy-
Szent-Miklós type, represents an independent alphabet."7

As specimens of this type of writing the following inscriptions have been

indicated by these authors: 1) two inscriptions on flasks from the Museum of
Novocerkassk; 2) five stone inscriptions from Majatskoe gorodisce; 3) rock
4
S. E. Malov, Pamjatniki drevnetjurkskoj pis'mennosti Mongolii i Kirgizii, Moskva-
Leningrad, 1959:68.
5
"[...] suscestvovanie dvukh sistem tjurkskoj runiki [...]", S. G. Kljastornyj,
"Drevnetjurkskaja civilizacija: diakhroniceskie svjazi i sinkhroniceskie aspekty" ST
1987,3:59.
6 S. G. Kljastornyj and I. Vásáry, "A Runic Inscription on a Bull Skull from the Volga
Region" in: Between the Danube and the Caucasus. A Collection of Papers Concerning
Oriental Sources on the History of Central and South-Eastern Europe, Budapest,
1987:171.
7
Kljastornyj - Vásáry, op.cit.: 172.
368 Edward Tryjarski

inscriptions from the ruins of Khumara (North Caucasus); 4) inscriptions

from Karakent (North Caucasus); 5) inscriptions on the Talas (Acyq-Tas)
stick; 6) a big inscription from Majatskoe gorodisce; 7) the inscription on the
bull skull being precisely the subject of their edition. They added that along
with these groups of the EER there still exist three other types of the script in
question, viz. 1) inscriptions from Murfatlar along with inscriptions from the
caves in the Crimea; 2) Avar inscriptions from the Carpathian Basin; 3) the
Székely script, "a late descendant of a local type of one of the EERs." 8 Quite
remarkable is the authors' following observation:

"Because of the lack of bilingual inscriptions and the insufficient number and
often fragmentary character of the documents, all attempts at their reading are
tentative, and for the most part improbable."9

Since the year in which these lines were printed archaeological finds and
other investigations have resulted in new observations and more or less sub-
stantial conclusions. It has become clear that the general situation in terms of
geography, chronology and ethnic policy is more sophisticated than had been
assumed, that some newly revealed writing systems were used by other,
mainly Iranian, peoples and that their anticipated Turkic appurtenance might
appear problematic. It has been supposed that the very repartition into Asian
and European runes might appear unserviceable, and even false, since both
groups might have possessed some deeper connections.
In recent decades, many new ideas on the runic scripts have been formu-
lated by G. Clauson, G. Doerfer, M. Erdal, V. G. Guzev, J. Harmatta, H. W.
Haussig, S. G. Kljastornyj, I. V. Kormusin, L. R. Kyzlasov, Gy. Németh, O.
Pritsak, A. Róna-Tas, O. F. Sertkaya, A. M. Scerbak, D. D. Vasil'ev and oth-
ers. Quite original, and perplexing at the same time, have been observations
and proposals made by I. L. Kyzlasov. With a solid grounding in archaeology
and palaeolography he was well-prepared to take a new look at old problems.
This should be emphasized since as soon as the golden age of Turkology,
marked by such names as W. Thomsen, W. Radloff, O. Donner, P.
Melioranskij or S. E. Malov, had come to an end, palaeographic studies on
the Turkic runes aroused no special interest. Only some time ago they were

8
Op.cit.:173.
9 L.c.
 The Unity or Multiplicity of Runic Scripts 369

again taken up by I. V. Kormusin, 1 0 O. N. Tuna 1 1 and D. D. Vasil'ev. 12

Lately, I. L. Kyzlasov has devoted much of his efforts to fundamental prob-
lems of the runic script. Sometimes one has the impression that his palaeo-
graphic pedantry goes too far and becomes a kind of I'art pour I'art13 but it
should also be remembered that the present state of research, characterized
by a richness and diversity of new materials on the one hand, and the fact of
our helplessness in the face of certain new finds on the other, simply de-
mands the creation of a dependable palaeographic base and the formulation
of bold new ideas. It has soon appeared that purely palaeographic observa-
tions have served I. L. Kyzlasov to formulate far-fetched hypotheses.
I. L. Kyzlasov started anew discussing some terminological questions
which threw his colleagues into confusion, as evidenced by a lack of any
broader acceptance of his proposals. He declared, for example, the traditional
term "Turkic runes" to be misleading and proposed to replace it by a "steppe
runic script" or "steppe runes". He named the script of five inscriptions of the
Ferghana Valley (viz. from the ruins of Kaladj-Kafir, Kaladj-Bolo, Kizil-
Piljau, Os-khona and Kuva) the "script of the Isphara" and the script of the
Nagy-Szent-Miklos treasure along with the inscription on the spindle from
Szarvas - the "script of the Tisza". In addition, he made a hypothesis - which
in most of his studies is presented as a proven fact - saying that in the basin
of the Middle and Upper Yenisei there had existed along with the generally
known "Yenisei alphabet" some other runic scripts, viz. a "South Yenisei
script" which up to the present day has been left unnoticed, and probably also
an "Upper Yenisei script," some traces of which have been observed by him

10
I. V. Kormusin, "K osnovnym ponjatijam tjurkskoj runiceskoj paleografii" ST 1975, 2:
25-47.
11
O. N. Tuna, "On the Phonetic Values of the Symbols , and . Used in Some of the
Texts in Kok-Turkish Script" CAJ 9, 1966,4:241-263.
12
D. D. Vasil'ev, Korpus tjurkskikh runiceskikh pamjatnikov bassejna Eniseja, Leningrad
1983.
13
"[...] these faults are not vital, there is enough reliable material to show what letters the
alphabets of these inscriptions contained, but no account should be taken of letters of
dubious shapes, particularly when they are parts of words which do not seem to make
sense, and it is sad that a good scholar like O. N. Tuna should have spent so much time
trying to find phonetic values for letters which probably never existed," Sir Gerard
Clauson, Op. cit:64.
370 Edward Tryjarski

and which can be tentatively ascribed to the Chik people. 1 4 The supposed
"South Yenisei script" must be related not only to the "Yenisei script" but
also, more closely, to the "Don script" and "Kuban script" as well. 1 5 In this
connection I. L. Kyzlasov proposes a new repartition and elimination of two
groups: an "Asian group" consisting of the alphabets of the Orkhon, the
Yenisei and the Talas, and "Eurasian group" embracing the "Don alphabet,"
the "Kuban alphabet," the "Isphara alphabet," "the alphabet of Acyq-Tas"
and the "South Yenisei alphabet." It should be understood that the old classi-
fication is rendered groundless and unnecessary.
It is interesting to know how I. L. Kyzlasov clears the way for his reparti-
tion. He cuts namely himself off all other runic type scripts the relation of
which to the Turkic world seems to him dubious, so not only off the Sekler
script but also off the inscriptions from the Black Sea shore, those of the
North-East Anatolia, those of the Balkan countries like the shamanic inscrip-
tion from Monastira near Ravna (tentatively but rather reasonably deciphered
by M. Moskov just as a specimen of an Asian runic script) 1 6 , that of the
Issyk Kurgan (resembling so much the Turkic runes that A. S. Amandjolov
did not hesitate to read it as a Turkic text) 1 7 along with similar texts from
Afghanistan. 1 8 I. L. Kyzlasov contends that the relation of those texts to the
"steppe runes" has not been proved. He writes in this connection what fol-
lows:

"Notwithstanding a widespread opinion concerning the affiliation of the

northeastern zone of the Black Sea to the Turkic runic script this cannot be really
proved. Studies by I. Donceva, E. Tryjarski, T. I. Makarova, S. A. Pletneva and
other scholars have demonstrated that the bulk of those inscriptions has a tamga-like

14
I.L. Kyzlasov, Drevnetjurkskaja runiceskaja pis'mennost Evrazii (Opyt paleograficeskogo
analiza), Moskva, 1990:117-128; the same, Runiceskie pis'mennosti stepnoj zony Evrazii.
Problemy istocnikovedenija. Avtoreferat, Akademija Nauk SSSR. Instytut Arkheologii,
Moskva, 1990:12-14; the same, Runiceskie pis'mennosti evrazijskikh stepej:42-56, 289-
320.
15
I. L. Kyzlasov, Runiceskie pis'mennosti evrazijskikh stepej:54.
16
M. Moskov, "Prabtlgarski runiceski nadpis" Palaeohulgarica - Starohhlgaristika,
1983:35-46.
17
A. S. Amandjolov, Tjurkskaja runiceskaja grafika III (nagljadnyj material - irtysskie,
18
ilijskie i syrdarinskie nadpisi), Alma-Ata. 1985:31-39.
A. A. Motamedi, "Discovery of an Inscription in an Unknown Language at Ai-Khanum"
Afghanistan, June 1980:45-48.
 The Unity or Multiplicity of Runic Scripts 371

character. In spite of a certain external similarity of a number of those signs to the

runes [...] neither single tamgas nor combinations of them appear to be actual texts.
[...] The published materials do not allow us to identify them with any runic
alphabet. The same problem applies to the inscription found in the village of
Ravna." 19

One gets impression that I. L. Kyzlasov has not been sufficiently in-
formed about all proposals lately made to explain the inscriptions from
Murfatlar and Pliska. 20 He seems also not to have seriously assumed his atti-
tude with regard to the Caucasian materials and proposals made by S. J.
Bajcorov in his book published in 1989. 21
His opinion on the well-know Kievan Khazar document is as follows:

"According to Pritsak, these are runes of the Orkhon type. His proposed
decipherment has been the result of some interpretations which are strained. [...] It is
clear that the inscription can be neither related to the Orkhon script, as proposed by
Pritsak, nor ascribed to any of the known alphabets. The signs number 1, 4, and 5
from the right make this impossible." 22

The above remarks have seemed necessary to draw your attention to the
scope and methods of I. L. Kyzlasov's research and, in particular, to his pro-
posals concerning the existence of the "South Yenisei alphabet". According
to him, first specimens of that alphabet were disclosed already in 1888, and

19
Cf. I. L. Kyzlasov, Runiceskiepis'mennosti evrazijskikh stepej:38.
2 0
Cf. E. Tryjarski, "Has a Key Been Found to Decipher the Eurasian Script of the Runic
Type?" in: Laut- und Wortgeschichte der Turksprachen. Beiträge des Internationalen
Symposiums Berlin, 7.-10. Juli 1992. Herausgeg. v. B. Kellner-Heinkele und M. Sta-
chowski, Wiesbaden, 1995:191, notes 13, 14, 194, note 30; the same, "Kritische
Bemerkungen über die neuen Versuche der Entzifferung der protobulgarischen
Inschriften" in: Turfan, Khotan und Dunhuang Vorträge der Tagung "Annemarie v.
Gabain und die Turfanforschung", veranstaltet von der Berlin-Brandenburgischen
Akademie der Wissenschaften in Berlin (9-12. 12. 1994.) Herausgeg. von R. E. Em-
merick, W. Sundermann, 1. Warnke und P. Zieme, /Berlin/ 1996:343-352.
2
' S. J. Bajcorov, Drevnetjurkskie runiceskie pamjatniki Evropy. Otnosenie severokav-
kazskogo areala drevnetjurkskoj runiceskoj pis'mennosti k volgo-donskomu i dunajskomu
arealom, Stavropol', 1989.
22
1. L. Kyzlasov, Runiceskie pis'mennosti evrazijskikh stepej:1A, 48.
372 Edward Tryjarski

from that year on 17 or 18 inscriptions in that script were brought to light and
published. They are as follows:

- 4 inscriptions from Sulek,

- inscription on a whorl of a spindle from the Minusinsk Museum,
- inscriptions on a vessel from Ujbat caatas,
- inscription from Sargol,
- 2 inscriptions on two horn-plates of the arc from Ajmyrlyq,
- inscription on the Ozemaja Mountain,
- 5 inscriptions on steles from Elegej,
- inscription on a stele from the cemetary of Turan,
- inscription from the Cinge River,
- inscription on a small arrowhead from Karban.

The list should be supplemented by another inscription from Turan lately

brought to light. 2 3
All of these inscriptions are short and represent (the Karban inscription
excluded) 43 signs. A remarkable feature of the "South Yenisei" inscriptions
should be a great heterogeneity of the variants of letters and word dividers.
One is allowed to suppose that those inscriptions do not present a palaeo-
graphic entity but belong to different specimens. The present state of South-
Siberian sources does not permit a well argumented analysis of the problem
and we must still await new materials. 24
According to I. L. Kyzlasov, 14 "South Yenisei signs" have no counter-
parts in the "Yenisei alphabet". In the case they have, I. L. Kyzlasov often
perceives a formal resemblance only. 2 5 Since the area of the "South Yenisei
script" must be located within the zone of the "Yenisei alphabet", the
possibility of the influence exerted by the latter on the former, or vice versa,
camiot be excluded. I. L. Kyzlasov's conclusion is that the proximity of both
scripts is, however, limited (to 11-12 signs), and he is rather unwilling to
speak about their genetic relationship. 26
His guess is that the "Yenisei script" and the "South Yenisei" script
coexisted for a certain time. Archaeological analysis regarding the "South

23
0p.cit.:289-320.
24
Op.cit.:43, 48-56 and Table XV.
25
Op.cit.:48.
26
Op.cit.:48-49.
 The Unity or Multiplicity of Runic Scripts 373

Yenisei" script allow to date its existence back to the second half of the 8th
century A.D. up to the 10th century A.D. This coexistence should be among
others evidenced by the text of Edegej I which is of mixed character, viz. is
written with the "Yenisei characters" but also contains three "South Yenisei
signs". 27
In connection with I. L. Kyzlasov's attempt at explaining the mixed char-
acter of some texts a brief remark can be made. It is reasonable to suppose
that the authors of those inscriptions had knowledge of both writing systems.
Certainly, they did not belong to broader social circles since knowledge of
reading and writing among the nomads of the epoch can hardly be supposed.
The question can rather be of skilled workmen, or experienced amateurs, who
were invited by neighbouring clans or tribes. In such circumstances a substi-
tution of one sign for another can be easily understood. On the other hand, it
would be interesting to dwell on some psychological reasons for the phe-
nomenon. We are allowed to suppose that they could be instilled in human
ambition, in a desire to mark one's individuality or ability. Such feelings
could be a feature of the engraver, of the ruler, of his kinsmen or representa-
tives. All of them would be happy to possess a useful, slightly cryptographic
means of communcation of their own.
An interesting and useful hypothesis regarding the existence of the
"South Yenisei" script demands, of course, the approbation of other
specialists. An attempt at verifying it is, however, rather difficult. The main
reason is that not all inscriptions forming a base for this hypothesis are
accessible in the form of photographs. This is the case of 11 or 12
inscriptions presented neither by earlier researchers like Malov, Batmanov,
Kiselev, Orkun, Vasil'ev, Kljastornyj, nor by Kyzlasov himself. With regard
to all analysed inscriptions the lecturer has at his disposal only copies
handwritten by I. L. Kyzlasov or by his predecessors. In some cases the
situation is delicate since the discoverer of the inscription, its copyist and its
editor are one and the same person, viz. I. L. Kyzlasov himself.
It is natural that editions of epigraphic monuments contain doubtful
points and misreadings; it is no wonder therefore that they also occur in the
texts studied by I. L. Kyzlasov. To provide a few examples, we might
indicate the inscription on a spindle in the Minusinsk Museum. D. D.
Vasil'ev (Korpus tjurkskikh runiceskikh pamjatnikov bassejna Eniseja,

27
Op.cit.:51.
374 Edward Tryjarski

1983:74, E 87) gives j , while the same sign is rendered by I. L. Kyzlasov

(Runiceskie pis'mennosti evrazijskikh stepej, 1994:299, Ju 5) as ^ (a sign
which supports I. L. Kyzlasov's hypothesis). In the case of Sulak VI, Kara
Jliz, H. N. Orkun gives the signs which seem to be retouched while I. L.
Kyzlasov most probably reproduces the present-day state of the monument
which lately has seriously deteriorated (the sign V ( p . reproduced by Orkun,
does not figure in Kyzlasov's edition at all). As concerns the Karban
inscription, I. L. Kyzlasov reproduces two foreign drawings, viz. one by V.
N. Jelin, and the other by E. P. Matockin. It is easy to perceive serious
differences in the proposed readings (J and M versus 3 and 1/1).
These cursory observations, the list of which could probably be length-
ened after a more detailed analysis, suggest that we should accept the new
proposals with caution. On the other hand, the existence of clearly written
signs such as Y , + , Q, f| ,) seems to speak in favour of I. L.
Kyzlasov's hypothesis.
The problem of the genesis of runic scripts, their mutual relation and
repartition demands further intensive research.