12 Mayıs 2016 Perşembe

Ancient Turkic Toponyms of the Middle Asia

A Scythian-Turk Warrior 
from Oxus (Oghuz) Treasure, 5th-4th c BC

Toponymic science firmly learned that territory of this or that people both in the present, and in the past can be determined with the help of toponymic areals. Toponyms are an extremely valuable source, a kind of “language of the land”, it gives frequently more for the solution of ethnogenetical problems than tens of fortresses or burials, than the inadequate and confused testimony of ancient and medieval authors, than studies of revered researchers. 

In the “language of the land” the ancient natives of territory had left stronger historical trace than great commanders who chiseled stone inscriptions. ‘‘Language of the land’‘ does not allow falsification. Geographical names are a type of documentary source inherently similar to archeological materials not dated and not deformed by time. 

However, archeological monuments are unfortunately mute: disclosing the material culture and even social relations of the past, they are silent about the languages of the founders of these cultures.

Like archeology, toponyms have a number of layers belonging to different periods of time. Determining the language of ancient names is not an easy task. A historian studying historical toponyms should treat with care the results of works of even recognized authorities in the linguistics, and should involve whenever possible all complex of the available historical and geographical data . 

Many place names recorded in the sources as Greco-Roman and Chinese names in the base have Eastern Iranian or Türkic origin. The toponyms given in the Chinese sources usually are calques or equivalents of the local Türkic or Iranian-lingual names . Precise etymologizing of these names demands from researcher even greater scrupulousness and all-round analysis.

Any place name exists not only from the moment of its recording in the sources. Hundreds, and maybe thousands years before the recording it could have been in the oral speech. Besides, far from all ancient toponyms were recorded in the written sources. One of the most ancient toponyms are the names of Iranian origin that survived in the territory of the Middle Asia 1* everywhere, both in the Turan plains, and in the Pamir mountains, but much less of them are in the mountain the areas of the Inner Tian-Shan, in the Jeti-su and the steppe areas of Kazakhstan. However, it is impossible to assert that the Iranian names are the most ancient in the region . 

There is a number of toponyms that can’t be positively attributed to establish their linguistic association 2*. In the southern zone of the northern hemisphere the initial toponyms began forming tens of thousand years ago . Is not also entirely true that only Iranists-Islamists are studying the historical geography of the Early Islamic period in the Middle Asia.

1* Under the term ‘‘Middle Asia’‘ we mean territory of the natural physical geographical Turkestan region limited in the west by the east the bank of Caspian Sea, in the north including Aral-Irtysh watershed, in the east - Tian-Shan, Pamir and Alai mountains, and in the south - Hindukush and Kopet Dagh mountains. In politico-administrative relation this term includes Uzbekistan, Tadjikistan, Turkmenistan, Kyrgyzstan, northern part of Afghanistan and a southern part of Kazakhstan.

2* Among these are Bukhara, Samarkand, Amul, Khiva, etc.

Certainly, the toponymy is, first of all, a linguistic science. However it does not belong to the linguists only. Etymologizing toponyms, alongside with the linguistic analysis, requires a study of the history, areal geography, and also specifics of the vision of localities and geographical conditions by the native speaking people. Solely linguistic approach, without taking into account historical and geographical data not always results in reliable ethymology. And in antiquity, and Middle Ages, like today, the toponyms was a research object of not only linguists, but also historians and geographers.

The materials of archeologicalal results do not allow to connect directly the Middle Asian civilization that arose in the 1st millenium BC with the cultures of the Bronze Age in same territory. The most ancient Pre-Indo-Iranian cultures in the Middle Asian interfluvial could be created by a traibal society and not necessarily connected with more complex forms of the social structure. It is known that since the 5th-4th milleniums BC, settled agricultural-cattle breeding communities with rudimentary proto-city cultures developed in the south of the Middle Asia, i. e. long before the coming of the Indoarian tribes.

Who were the founders of these most ancient cultures of the Middle Asia is not established till now. Supposedly, before the Indoarian occupation of the Middle Asia in the middle or the second half of the 2nd millenium BC, its southern areas were peopled by carriers of the Dravidian language. It is known that Dravidizms make a significant part of the Sanskrit dictionary.

The newest toponymic and linguistic research allow to posit that in the 2nd millenium BC carriers of the Dravidian languages were in immediate contact with the carriers of proto-Türkic languages, however their connections were severed by the invasion of the Indoarian newcomers . In some Türkic languages in the European part of Russia (Bulgarian, Khazarian, Chuvash), Siberia (Yakut) and the Far East survived the traces of the proto-Türkic language widespread in the most ancient times in the southern part of the Middle Asia. These relicts point to the presence in the remote past of the ancestors of the carriers of these languages in the south of the Middle Asia, and to their subsequent migration to the north. These linguistic data can be compared with some data of the arheological studies that found cattle breeding tribes in the middle of 2nd millenium BC, the ancestors of carriers of the Karasuk cultures in the Minusinsk depression, who came there from the Middle Asian steppes.

The Türkic languages retained numerous traces of the most ancient linguistic layer, common for Middle Asia and Caucasus going back to the primitive-communal system epoch. These traces retained in those parts of the speech which belongs to the basic linguistic fund of the language (for example, the names for parts of human body) and as a rule are not borrowed from other languages. In the linguistic literature the most ancient forms of the Türkic languages are called to proto-Türkic, their carriers are called pra-Türks, and the history of these languages is closely connected with the family of so-called paleo-Asian and Enisei languages. Paleo-Asian and Enisei languages, according to toponyms, before the end of the 1st millenium BC covered a significant part of the Middle Asian territory, and most actively participated in the formation of the nucleus of the Altaian-Hunnish people conglomerate.

According to another linguistic theory, the carriers of the so-called ‘‘proto-Tokhar’‘ language, a most ancient of all Indo-European languages, in the 2nd millenium BC crossed the Middle Asian territory on the way from the Near East to the Eastern Turkestan, and contacted there with the carriers of the early proto-Türkic, proto-Ugrian, and Enisei languages. Plausibly, the interaction was long enough, and probably brought bi-linguality between the contacting groups of the population. This linguistic theory recently found convincing confirmation in archeological materials.

The names of Türkic origin spread in the territory of the Middle Asia from the extreme antiquity constituted a significant part of the medieval names, and continue to be an overwhelming part of the modern toponyms in this region. During early Middle Ages epoch they were a second in number after the toponyms of the Eastern Iranian origins, and in the Middle Ages they constituted the most powerful toponymsc layer in the Middle Asia. Only in the territory of the modern Kyrgyzstan were located about 300 ancient place names known in the historical sources since antique time and till the 12th century. 

The Türkic lexical contribution is recorded in many languages of Europe and Asia, including toponyms. Therefore the etymological study of toponyms in this huge region, like in the Middle Asia, requires application of the Türkic linguistic sphere. In the Balkans, and in the Eastern Europe, many geographical objects and settlements had parallel Türkic names. The same also applies to the Middle Asia, where many cities and settlements had two names, Sogdian and Türkic, and some geographical objects had only one, Türkic name.

The beginning of the active formation of the Western Iranian (Persian) toponyms in the Middle Asian territory was the period of the Arab conquest. At that particular time came into use new Persian topoformants, -dekh and -abad with a meaning of ‘‘settlement’‘, together with the New Persian language (Farsi) of the Western Iranian origin they spread in the Middle Asia. During that time in the Middle Asia appeared toponyms with the Western Iranian topoformants -an , -kird (-gird, - djird), - diza (-diz) , -rud and -stan (-istan). The actual Arabic elements in toponyms are insignificant and are mainly found in the south of Uzbekistan and in the Zarafshan valley. The beginning of the Persian penetration into the Middle Asia can be taken as 21/642, when the Sasanid army suffered a defeat from the Arabs at Nihavand, and the Persians had to flee across the Amu Darya.

In the last decades, because of intensive development of the archeological, numismatic, historical, linguistical and source, and accumulation of huge amount of materials across the Middle Asia, the historical geography also received a big boost, including in the toponymic science. Today, a time has already came for complex studies using mass of information from the various historical disciplines.

Many (but far from all) works of the Middle Asia philologists in the area of etymological studies suffer weak argumentation because of insufficient source base and scientific literature. On the other hand, many European researchers also can be accused in biased views and unwillingness to consider opposite opinions. A. Z. Validi Togan wrote that according to the obsolete views, Bartold and Tomashek stipulated that during pre-Mongolian period in the Maverannahr were no Türks that transitioned to cultural life. For that reason by all means they tried to read Türkic names in Maverannahr only in Sogdian and Persian. Research of historical toponyms should evaluate all available data, including the works of the Middle Asian philologists who, as local residents, can analyze not only the data of the historical linguistics, but also many other factors, such as landscape and local features, folk legends, and idiosyncratic view of the territorial history by the local population.


The most ancient of all names of the Middle Asian region is "Turan", with the root of ethnonym Tur, a general tribal name of the ancient nomadic and semi-nomadic peoples of the Middle Asia. In beginning of the 1st millennium BC the Turs were a large military and political association antagonistic to the agricultural oases of Iran. This ethnonym, for the first time recorded in the "Avesta", subsequently became a root in the toponym Turan, mentioned in the ancient Iranian mythology [Firdousi] and Middle Persian religious and historical literature and documents , and also in medieval Persian, Arabo-Muslim and local Middle Asian historical and geographical literature.

The epoch of the Sasanid king Shapur I (ruled in 242-272) the name Turistan came into use among the Persians, it gained distributed alongside with the name Turan and had identical meaning the “country of Turs”. The ruler of that country is Sacan-shah, i. e. “Saka King”. On the Sasanid seals of the 3rd-4th centuries alongside with the names Turan and Turistan also appears the name Turgistan as their equivalent. Later, during the Ephtalite (5th century) and Türkic (6th-8th centuries) epochs, it transformed into the present form "Turkistan" (or Turkestan), which means “country of Türks”. The toponym Turan (trgn) is recorded on the Khoresm coins of the early Kushan and Early Middle Ages time. The name Turkestan, i. e. the “country of Türks”, in the perception of the ancient Iranians geographically covered the territory of the basins of the rivers Amu Darya, Syr-Darya and Tarim.

The equivalence of the ethnonyms Tur and Turk (Türk), like that of the toponyms Turan and Turkestan, first of all took place in the Iranian-lingual historical-geographical tradition of the pre-Islam and Islamic time , and then in the Arabo-Muslim and Middle Asian literature. The ancient Iranians called Turs first of all the Middle Asian Türks, and also the Türkic-lingual Sakas and Huns. From the 6th-7th centuries the Turan geography assumes quite real borders, which coincide with the borders of the Türkic Kaganate, and include the lands located south from the Amu Darya. In the "Shahname" the city Samangan is called “possession of the Türkic Kagan”.

In the 11th century the Türkic authors identified the legendary king Afrasiab of Turan with the legendary Türkic Hakan Alp Er Tonga. The name Alp Er is mentioned in the ancient Türkic runiform inscriptions from the Altai mountain. The fate of the Chionite king Ardjasp, described in the Parthian epos “Ayatkar Zareran”, coincides with the Türkic legend about the Türkic ancestor Ashina. The ethnonym Turk or Türk is a two-element (Tur-k) composition, where the second element "-k" is an affix of collective plural, and as a whole this word is the Türkic form semantically equal to the Iranian Tur-an.


The name ‘‘Maverannahr’‘ (Ma vara'an-nahr), which in the Arabic means ‘‘that behind the river’‘. Under the name ‘‘that behind the river’‘ (Ma vara'an-nahr) the Arabs meant the country of settled Türks in the Middle Asia.

The term ‘‘Maverannahr’‘ for the first time is found in the hadises about the prophet Muhammad, it probably was spread among the Arabs still in the pre-Islamic time, when it was understood as the lands located northeast from the Sasanid Iran. The Arabs who were making their first campaigns against these lands distinguished two concepts, ‘‘Ma duna-n-nahr’‘, which means ‘‘that in front of the river’‘ and ‘‘Ma vara'an-nahr’‘which means ‘‘that behind the river’‘ where under the ‘‘river’‘ in this case was meant Amu Darya. 

The largest river in the Middle Asia, Amu Darya, mentioned in the ‘‘Avesta’‘, was called Vahvi Daitya, the ancient Greeks called it Oxus and identified it with the name of the river Vakhsh. The Türks simply called it Okuz, i. e. ‘‘river’‘ because the ancient Türks called any big river okuz, and this word also has a second meaning ‘‘bull’‘.

The initial names of the rivers “water” in the mythological vision was connected with various images of tribal animals, and first of all with the totems. Supposedly, the ancient Greeks encountered the name Okuz and transformed it into Oks (Oxus) or Akes. The population of the Middle Asia untill the 17th century. called Amu Darya Asaf-Okuz or simply Okuz.

On the Catalan map composed in the 1375 from the information of the eyewitnesses, is shown a city Ogus, located near the mouth of the river Amo (i. e. Amu Darya). On the 1562 map of Antonio Djenkinson the lower course of the Amu Darya running into the Caspian Sea was called Ugus (Ougus). On some maps of the 16-17th centuries Amu Darya was called Ugus that means ‘‘bull’‘. 

During the Middle Ages, the name Okuz also had the Binakat valley and the city Iki-Okuz located between the deserts Ila and Yafindj. The river near the city Osh in the Fergana valley was called Tavushgan-Okuz. In the Tokuz-Oguz country, the city Irguzkat, with name’s etymology coming from the Türkic Ikki Oguz which means ‘‘between two rivers’‘. On the bank of the Aral Sea is mentioned a mountain Chagyroguz (Djagiragur). In the East Turkestan are mentioned the rivers Ikki-Okuz, Tumushgan-Oguz, Kara- Kash-Okuz, in Mongolia are mentioned Orkun-Uguz, Togla-Uguz and Yar-Uguz. 

This word was also preserved in the the name of the Uzboi (Okuz-Boy) dry channel, which flows into the Caspian Sea, and also in the diminutive form Ozek (Okuz-ak), which is used for the small rivers, in the hydronyms Kok-Ozek, Kyzyl-Ozek, Sary-Ozek (Blue Rivulet, Red Rivulet, Yellow Rivulet - Translator’s Note), etc. . The name of the city Uzkand (Uzgand, Ozd-jand) in the Fergana valley, also is probably formed with the topoformant okuz/uz/oz with the meaning ‘‘river’‘. 

The richest toponyms in the ancient Türkic is connected with the names of mountains, to designate which the ancient Türks used words as tag and art.


The greatest distribution in the Middle Asia have oronims with topoformant -tag (-dag) or-tau (-tuu) based on the ancient Türkic word tag ‘‘mountain’‘. In the modern geographical nomenclature this word is used for designation of separate mountain or a mountain ridge. 

The toponyms belonging this group have been recorded, for the first time, in the ancient Greek sources. 

"Megabazos bu işlerle uğraşırken, bir başka önemli harket da Libya'ya karşı açılmıştı, önce bazı açıklamalr yapacağım ve nedenini sonra söyleyeceğim. - Brauron'da Atinalıların karılarını kaçıran Pelasglar, Argo gemisi denizcilerinin üçüncü kuşaktan yeğenlerini Lemnos'tan kovmuşlar, onlar de deniz yoluyla Lakedaimon'a gitmişler. Taygetos üzerinde kamp kurmuşlar, ateş yakmışlardı....(Heredot:Müntekim Ökmen,1991)

Herodotus [4. 145] 

"About this very time another great expedition was undertaken against Libya, on a pretext which I will relate when I have premised certain particulars. The descendants of the Argonauts in the third generation, driven out of Lemnos by the Pelasgi who carried off the Athenian women from Brauron, took ship and went to Lacedaemon, where, seating themselves on Mount Taygetum..." 

[....they sailed away to Lacedaemon, and there camped on Teügetum and kindled a fire...link]

[....and sitting down on Mount Taÿgetos they kindled a fire...link]

In the Pehlevi composition ‘‘Shahristanha-i Iran’‘ (Cities of Iran) is mentioned mountain Ek-tag (Golden Mountain) where stood headquarters of the Türkic Kagan. The Byzantine sources are mentioned oronims Ek-tag and Qaz-tag. In the Kyzyl Kum desert are relic mountains Djumurtau with name formed from the ancient Türkic word djumur and topoformant -tau and means ‘‘Round Mountains’‘. In the northern Khoresm is a mountain Kubatau with name formed from the ancient Türkic word cuba and means ‘‘Flat Mountain’‘. To the same category also belongs the oronim Ala, which endured till now in the form Ala-tau.

Among Türkic peoples Tian-Shan mountains were called Tangri-tag, meaning ‘‘heavenly mountains’[Tangri-God-SB]. The Chinese name Tian-Shan also means ‘‘Heavenly mountains’‘ and is a calque of Türkic name Tangri-tag. Sima Quan wrote that Huns called mountains Bai-shan (i. e. Tian-Shan mountains) ‘‘Heavenly’‘, from which follows that Türkic name Tangri-tag was in the use by the ancient Huns. The name Bai-shan, like and Sjue-shan (Snow mountains), Possibly also considered as a calque the local name Karlyk-tag (Snow mountains). The absence of the Iranian calque of name Tangri-tag allows to assume that these mountains were initially known with their Türkic name, from which subsequently was copied the Chinese name Tian-Shan. [Tai Shan -Tai Mountain-SB]

Names of many mountains are connected with mineral deposits. So, in the medieval sources are mentioned such oronims, as Komurtag (Coal mountain), Demur-tag (Iron mountain), Altun-tag (Gold mountain), Altun-kan, Bakyrlyg-tag (Copper mountain), Goshun-dag (Lead mountain), Kumushkan, etc.

Professor Mirsodik Ishokov
Editor-in-chief Doctor of Historical Sciences, 
Tashkent, Shark, 2006, to read more:

Also read:


Difference between Persian, Scythian and Hellen

Scythians were not İranian (Persian) speaking people.
Like in these example, in the 5th century BC their dress is not in Persian-style. Persians don't wear trousers, fur and 
so-called "Phyrgian Hat". This style of dress belongs to nomads, to Scythians and Sacae people, who lives on horses, 
fights on horses, sleeps on horses and dies on horses.

Ancient writers never wrote that they spoke Persian.
"The Royal Scythians were the ancestors of the Türkic speaking peoples." 
(P.I.Karal'kin,1978, a Tuvan ethnographer)

"If the Scythians, Sarmatians and Alans were Ossetian speaking, all Eurasia should have Ossetian toponyms. 
They do not exist, unless artificially (quasi-scientifically) produced. Thus, in all their attributes the Alans were Türkic, 
and took part in the formation of the many Türkic peoples."

Mirfatyh Zakiev
Collection of articles on problems of lingohistory, revival and development of the Tatar nation
Kazan, 1995