The Mushki; a lost Armenian tribe?

Guest entry – Author: Norgrmaya

Russian historian Igor M. Diakonoff (in)famously suggested in his oft-cited 1968 work, “The Pre-history of the Armenian People” that the Armenian language was brought to the Armenian Highlands in the early 12th century BCE, following the Bronze Age Collapse (when the Hittite Empire fell, leaving a power vacuum in Asia Minor and the eastern Mediterranean). Diakonoff theorized that the original speakers of Armenian are to be identified with the Mushki, a people attested by the Assyrians, and later in Vannic inscriptions, as well as by the Luwians, and possibly Greeks.

According to Assyrian records, the Mushki, with their allies the Urumu and Kaska (sometimes called Gagas, Kaskians, or Apishlu) attempted to invade Assyria from the north around 1165 BCE, taking control of Purukuzzi and Alzi (which may have been a variant of the name “Azzi”). These Mushki were retroactively coined “the Eastern Mushki.” Apparently, after defeating the Mushki, the Assyrian king Tilgrath-Pilesar I resettled many of them in northern Syria.

In the 8^th century BCE, a group called the Mushki were mentioned in the annals of the Kingdom of Van (a.k.a. Urartu) and Assyria as being present in Tabal, near Cilicia. This group were later dubbed “the Western Mushki.”

The 5th century BCE Greek historian Herodotus said:

“the Armenians, who are Phrygian colonists, are armed in the Phrygian fashion,”

and stated that the Phrygians originally came from Macedonia. Diakonoff, using Herodotus’ statement and the fact that the Assyrians seemed to connect the Mushki with the Phrygians and Phrygia, asserted that the Mushki were either the same as the Phrygians, or a closely related group who followed them from the Balkans eastward toward Armenia. Confusingly, Diakonoff (and other scholars) also connected the Mushki with the Moschoi attested by the Greeks, even though the Greeks differentiated between the Moschoi and the Phrygians. The ancient Greeks (such as Hecateaus of Miletus) placed the Moschoi in the Pontic region of modern northeastern Turkey and possibly also near Lake Urmia in Iran.

Additionally, Diakonoff and others further suggested that the Meskheti/Moschians of Javakh in Georgia were also Mushki, leading some to suggest that the Mushki spoke a Kartvelian language.

Unfortunately, the Mushki left no records of their language. The only personal name that is associated with them is the 8th century BCE (Western) Mushki king, Mita. Mita, who fought against the Assyrians, is speculated to be Midas of Phrygia (located in modern Cappadocia in central Turkey). This Midas is thought to be the inspiration for the mythological king who turned everything that he touched into gold.

However, modern research tells us that much of Diakonoff’s argument is untenable. Genetic studies have determined that the majority of the Armenian ethnogenesis was completed by 1200 BCE. This suggests that if the Mushki were from the Balkans, they could not have contributed to the Armenian genepool as their entrance was too late and there are no Balkan-origin markers in Armenians dating to the period in question.

Linguistically, the Phrygian language (which we do have records of, unlike whatever the Mushki language was) is now believed to be closely related to Greek.

Despite these issues, Diakonoff’s proposed Armenian-Mushki connection is actually fairly compelling—when one recognizes its major flaw: Diakonoff likely got the direction of their migration wrong.

There is no archaeological evidence of the western (so called “Phrygian”) ceramic ware reaching beyond Cappadocia in the Bronze Age. However, there is evidence of ceramics from the east reaching as far into the interior of Asia Minor as modern Elazig during the time in question (roughly 1200 BCE). This pottery style is descended from the Middle Bronze Age “Transcaucasian” ceramic ware, originally found in southern Georgia and northern Armenia, which spread throughout much of modern Armenia and as far west as Erzurum by the middle of the 2nd millennium BCE. The Transcaucasian ceramic ware is tied to the Trialeti-Vanadzor Culture, which both archaeological evidence and now genetics strongly hints at having been an Indo-European Culture. It is very possible that the Trialeti-Vanadzor Culture spoke the Proto-Armenian language.

According to historian Aram Kossian, the Elazig ceramic ware’s distinctiveness from the previous ceramic styles found in the region is suggestive of a new population—one from the South Caucasus and Armenian Highlands—settling in the Elazig region at the end of the 2^nd millennium BCE. Additionally, Kossian argues that the large amount of this pottery found in Elazig could indicate a 50% increase in population. Kossian thinks that this point to the settlement of a population originally from the South Caucasus.

In other words, the archaeological evidence partially refutes Diakonoff’s theory: it points to an eastern (Armenian or the vicinity) origin of the Mushki and not a western (Balkan) origin. Kossian concludes that the Mushki were either speakers of Armenian, or speakers of a closely related, although now extinct, language.

Speculation that the Mushki may have lent their name to the city of Mush as well as the ancient region of Moks (Moxene in Latin) could have some validity as the Vannic inscriptions placed the Mushki tribe in southeastern Asia Minor.

Diakonoff partially connected the name Mushki to an Armenian origin, stating that the root of the name Mushki was some form of “Mush/Mus/Mos/Mosh” (a word he didn’t etymologize) with the addition of the Classical Armenian k’ plural suffix (as we still see in words, such as Barsik’ (Persians, or literally, Parsis)). Armenologist Armen Petrosyan extrapolated on this idea, positing that the ki suffix could be an earlier, Proto-Armenian form of k’. Unlike Diakonoff, however, Petrosyan offered an etymology and definition for “Mush,” saying that it could be related to the Armenian “mshak” (vineyard worker, farmer) and/or Kartvelian “musha” (worker).

Some sources suggest that Mazaca (now called Kayseri) was the Mushki capital city. Interestingly, the medieval Armenian historian-monk Movses Khorenatsi (Moses of Khoren) wrote that the mythological Armenian patriarch, Aram, made his cousin/general, Mishak (Missak), governor of Cappadocia. According to Khorenatsi, Mishak founded the city of Mazaca and named it after himself.

In addition to the name Mishak corresponding with Mushki, the name Aram could correspond with the name of the Mushki’s allies, the Urumu. In cuneiform writing systems, vowels are interchangeable. It is possible that Urumu was actually pronounced Aramu or Arami or some variation (incidentally, this is similar to the name of the first king of the Kingdom of Van, Aramu/Arame).

According to 1st century CE Jewish scholar Flavius Josephus, Mesech (Mosoch) was the father of the Mosocheni, who later became known as the “Cappadocians.” In Genesis, this Mesech is a son of Japeth. A possibly different Mesech (known as “Mash” in Genesis) was mentioned as a son of Aram and brother of Hul in Genesis. Josephus stated Hul founded the Armenian nation. It is worth noting that Hul could theoretically be rendered as “Hur” or “Har” (which could be connected to both the names “Hur[rian]” and/or “Hark” (“the Ar people”) and the primary Vannic deity, Haldi).

It is often suggested that Mita, the only surviving personal name associated with a Mushki, is of Phrygian derivation due to its association with King Midas of Gordion, which was the capital of Phrygia. However, the name Mita was recorded as early as the 15th century BCE in relation to the king of Pahhuwa, a Hittite vassal state located near the Upper Euphrates in or near Hayasa-Azzi. If the Phrygians and/or the Mushki emigrated to Asia Minor from Europe after 1200 BCE, there is no way that Mita is a Phrygian name. It is possible that the name Mita is indeed Armenian, deriving from “mit” միտ (meaning “mind,” “intellect,” “understanding”), which ultimately derives from Proto-Indo-European *med. One of the kings of Mannaea, an eastern neighbor of the Kingdom of Van (Urartu), located near Lake Urmia, was named Mitatti. This name could be related to Mita, perhaps meaning something like “wise god” (“ti” դի being “god” in ancient dialects of Armenian, a cognate to Anatolian: *diu and cognate to Latin: *deus, all from Proto-Indo European *deywós).

Another option is that Mita derives from “mitta”/“miti,” denoting a red wool used for Hittite ritualistic purposes.

Maybe Mutti, a name mentioned by the Hittites in relation to Hayasa-Azzi, was a dialectical variant or Hittite rendering of Mita, although it’s also possible that the names are unrelated.

Diakonoff was adamant that Hayasa-Azzi had no relationship with Armenians. However, confusingly, he argued that the Chaldians, a people living on the southeastern Black Sea coast in the same geographic region as Hayasa (albeit some centuries later), could be connected to the (Eastern) Mushki (who he considered to be the Proto-Armenians).

The Cambridge Ancient History states that the (Eastern) Mushki may have contained Hurrian and Indo-Iranian elements based on personal names recorded in the eastern Euphrates at the time of Mushki presence. While these names are not explicitly connected to the Mushki, if the individuals bearing these names were indeed Mushki, it would affirm their eastern origin. Both Hurrians and Indo-Iranians (Mitanni, possibly Kassites) were neighbors of Armenians. Is it possible that some ethnic Armenians bore Hurrian names? Or similarly, that some Armenians had Indo-Iranian names (as many Armenians do today) or even that some of these names are both Armenian and Indo-Iranian due to the shared origins of both groups?

Personal names attested between the 12^th-9^th centuries BCE in Tabal (Cilicia) and Bit Agusi (northern Syria), such as Arame/Aramu (who shared a name with the first king of Van) and Armanani, bear a striking resemblance to Armenian names. While records in these regions were written in Luwian (an Indo-European language closely related to Hittite) and Aramean, could these names be evidence of an early Iron Age Armenian presence or influence in this region, perhaps via the Mushki (who had a presence in the region, according to the Assyrians and Vannic inscriptions)? Tabal sided with the Kingdom of Van over Assyria. Perhaps this was due to cultural ties?

Diakonoff’s theory that the Mushki were connected to Armenians seems to have been somewhat correct, however, many elements of his argument are not supported by the evidence available today. There is no reason to believe that the Mushki migrated from the west at the end of the 2nd millennium BCE. In fact, there is little reason to believe that the Phrygians moved into Asia Minor at this time either, besides notes written by the Greeks more than 600 years after the Phrygians were supposed to have arrived in Cappadocia. This has left some scholars to posit that the Phrygians were natives of Asia Minor themselves. If that’s the case, then it is possible that the Phrygians (as the Mushki) may have come from Armenia and moved westward into Cappadocia and Phrygia.

The other option is that the Mushki established themselves as a ruling class over the Phrygians or vice versa (which could explain the Mita/Midas connection) and the Greeks started to equate the two groups with one another. Or perhaps the Greeks confused the Mushki (as Moschoi) with the similarly named Mysians (Musos) who lived in western Asia Minor and seem to have been connected to the Phrygians linguistically.

While the nature of the relationship between the Armenians, Mushki, and Phrygians is still unclear, one thing is certain—there is no reason to assign the Armenians an origin outside of the South Caucasus, despite Diakonoff’s insistence to the contrary.