The Narasimhan et al. paper disproved the predictions of “Swat Aryans and Achaemenids being 70% like Sintashta”. The original conclusion of the paper was that while the steppe ancestry in Swat was mostly mediated by females there was a high steppe migration into the Punjab that brought Indo-European languages. On the other hand, there was nothing on the Iranian migrations other than the DA382 sample with significant MLBA steppe ancestry which is erroneously dubbed by some as “Proto Iranian” or “Early Iranian”. Not only is it impossible for him to be a Proto-Iranian as he is dated to 850 BCE when west Iranians had already split and migrated to the Zagros and formed kingdoms on the borders of Assyria but he is also too late for the Early Iranians that appear in the Rigveda as generally accepted (Iranian names occur in books 2, 6, and 8 of the Rigveda as per Witzel (1995) and there is an attested west Iranian name in west Asia even centuries before. A while after I was done writing the first draft the “Hybrid Model” came out which made Narasimhan sit on the fence when it came to Indo-Iranian.

The Kulturkugel models that are proposed mainly by Mallory and Parpola, where an Indo-Iranian group either forms or adopts the BMAC culture but keeps its language, and the different migration models supported by Kuzmina, Witzel, and others both contradict the current data.

For the Kulturkugel model, Parpola (2015) sees the horse scepters in BMAC as evidence of steppe warrior-elite. Fortunately, we have a sample of a warrior buried with a horse scepter, “Gonur Tomb 2380 sample 17 (I1784): Date of 2201-2031 calBCE (3720±30 BP, Poz83485). Genetically male. Nicknamed the ‘Tomb of the Warrior,’ this was skeletally a male, 40 to 50 years old at the time of his death, flexed supine and oriented north-northwest. The neck of the man was broken, and this was the probable cause of his death. He was buried in a shaft grave on the southeast edge of the large cemetery. This very rich grave was accompanied by one bronze knife, one silver plate, one bronze vessel (diameter 16 cm and height 12 cm), one bronze mace head in the form of a horse head, one bronze mace head with four spikes, one bronze semi-cylindrical artifact near the head, one bronze 118 leaf-shaped arrowhead near the pelvis, and one bronze plate with perforations wrapped in linen cloth near the right shin”, of typical Geoksyur Y-DNA J1a2a1 (also carried by the Hasanlu outlier) and MtDNA U7. In fact, the difference between bronze age BMAC and the early copper age Turanian (Turan, as named by Harvard, being the historical Khorasan region) samples is simply an increase in West Asian and Indus-related ancestry which is no surprise as BMAC is believed to be formed from the earlier Namazga culture with Mundigak influence and prolonged contacts with IVC. Quick models of I1784 below:

Target: Turkmenistan_Gonur_BA_1:I1784
Distance: 2.0984% / 0.02098413 | R3P
61.4 Turkmenistan_C_Geoksyur
25.2 Iran_C_SehGabi
13.4 Iran_ShahrISokhta_BA2

Target: Turkmenistan_Gonur_BA_1:I1784
Distance: 1.8950% / 0.01895015 | R3P
80.6 Iran_C_TepeHissar
13.8 Iran_ShahrISokhta_BA2
5.6 Kazakhstan_Dali_EBA

Meanwhile, the migration model fails in every respect. In Iron Age Swat Valley, the gateway to India, there are only two samples with R1a, I12457 (970 BCE) and I12450 (800 BCE), and two samples with a peculiar I2a dating to 900 BCE with most of the steppe ancestry being mediated by females. There is even less evidence for a migration in late/post-BMAC (1600-1300 BCE) where according to Kuzmina there is clear evidence of a “Fedorovo fire cult” in Bustan, which is described in the supplementary materials of Narasimhan et al: “Archaeological investigations at Bustan Burial Mound have revealed a complex funerary ritual related to the usage of fire. On top of the graves there were piled rocks, showing the influence of Steppe traditions. There were inhumation as well as cremation burials. There was a dedicated chamber for cremation of bodies at Bustan, including multi-usage hearths and altars. The altars were functionally classified into ones used for libations, ones used for meals, and ones used for sacrifices. The funerary rite documented at Bustan, specifically in relation to the role of fire, is not known at this time from any other site Iran, South Asia, or the Central Eurasian Steppes”. The Bustan samples are simply a continuation of Gonur as Alberto has previously shown:

So, in the migration scenario females with steppe-related ancestry move down across the IAMC by the end of the bronze age and this leads to the formulation of the “patriarchal-steppe-warrior-chariot” bronze age hymns somehow in iron age India? And Iranians all go back to post-1000 BCE Yaz II people like TKM_IA, who probably descends from the Kuyisay Saka that reached Khwarezm 1200 BCE, yet somehow manage to compose the Avesta before the Iranians migrate to western Iran and somehow the Medes and Persians separate around 1000 BCE. Also, by some miracle, the Mitanni reached West Asia around 1760 BCE. This is simply impossible.

This should not be surprising: archaeologists for years found little evidence of a movement from Andronovo to India and Iran, as one of the godfathers of the steppe hypothesis Mallory (1998) once stated: “this type of explanation only gets the Indo-Iranian to Central Asia, but not as far as the seats of the Medes, Persians or Indo-Aryans”. The Andronovo model was always challenged by archaeologists, Jarrige and Hassan, and many others including Lamberg-Karlovsky, who thought that there were nationalistic motives behind this model and that Andronovo could have spoken a non-IE language (Lamberg-Karlovsky, 2002). In my opinion, it is very clear which direction the less dogmatic proponents of the steppe hypothesis swing to, which is the BMAC-Kulturkugel route, yet BMAC itself does not show any evidence of a steppe takeover. Nevertheless, these scholars do not deny that many of the characteristics and markers associated with the Indo-Iranians are in fact “native”. For example, Erdosy (1995) noted that cremation was pretty common in 2000 BCE Balochistan (Penano Ghundai II, Mughal Ghundai III, Dabar Kot, Mehi, Sutkagen-dor) but very rare in Central Asia: “If anything, on present evidence, cremations appear to have originated in the Indo-Iranian Borderlands and spread northwest (and southeast) thence, against the grain of postulated movements of Indo-Aryan speakers”. It’s interesting how the Andronovo scenario was never doubted even though the Sogdian hypothesis of Nichols and the Armenian hypothesis of Ivanov and Gamkrelidze were considered fringe theories because there was no archaeological evidence at the time.

However, thanks to aDNA we can now trace the movements of the Indo-Iranians to the steppes with Kyzulbulak_MLBA2 and more importantly the J2a1a4b Taldysay_MLBA2 (~1500 BCE). This haplogroup probably made its way from Hissar (the oldest sample is I2337 from 3500 BCE) to Turan as it is found in BMAC-related sites and Swat. There is also a link with later steppe nomads such as the MJ-16 Scythian (600BCE) in Ukraine and the Tian Shan Saka.

We also have BMAC-related pottery appearing in the steppes, “In the 2nd millennium BC settlement of Shagalaly II, between Kokshetau and Zerinda, in the Akmola region of northern Kazakhstan, T.S. Maljutina brought to light imported pottery of Oxus origin, or imitation, in stratigraphic association with Fedorovo and Alekseevka pottery”. The BMAC pottery continues well into the post-BMAC period, “A baked-brick hypogeum dated back to the 14th–13th centuries BC was excavated by V.K. Mertz in the graveyard of Karaoba, along the Irtysh River valley, south of the Burly archaeological facies… It contained some wheel-made pottery sherds belonging to large jars of the late Oxus Civilization production in close association with numerous Begazy-Dandybaj complete vessels… This funerary structure is the earliest and easternmost baked-brick hypogeum excavated thus far in the steppes of Eurasia. Other finds of wheel-made Oxus pottery have been documented at Begazy-Dandybaj sites in central Kazakhstan: Myrzhyk… Kent, Domalaktas, Bajshura, and the Tasyrbaj burial ground… Similar finds are known in the Inner Syr Darya River delta, in the Tagisken 2 burial ground, where ceramic vessels of southern Central Asian production stratigraphically are associated with funerary hypogeums built in baked bricks” (Bonora, 2020). There is enough archaeological and genetic evidence at the right time for a BMAC-related group to bring Indo-Iranian languages to the steppes.

Indeed, with the current data we reach the same conclusion that was hinted at long ago by Jarrige and Hassan (1989): “We leave to the linguists the problem of whether Indo-European languages were introduced into the Middle Asian regions from a still unknown part of the Eurasian steppes in the course of the third millennium or if Indo-Iranian languages have been associated with these regions for a much longer period. As far as archaeology is concerned, we do think that it is increasingly necessary for specialists in Indo-lranian studies to pay attention to the . . . interrelated cultural entities of the late third and early second millennium in the regions between Mesopotamia and the Indus. It is a direction of research that is likely to be more fruitful than are traditional attempts to locate remains left by nomads from “the Steppes,” attempts that were in fashion when the Indo-Iranian Borderlands were thought to be a cultural vacuum”. As I will argue below, even before aDNA BMAC with its Varuna palaces, fire altars, soma cults, three-headed dragons..etc, it shows one of the clearest examples of Indo-European religions. As such, the point of this post is to disassociate the Proto-Indo-Iranians and the steppes and revisit the available data to offer a better, more comprehensive model in which Proto-Indo-Iranian reaches South Central Asia in the Late Copper Age.

 

Pseudo-Aryans of Andronovo

Before I dive into anything I have to explain my gripes with the whole Andronovo scenario other than the fact that there is no genetic nor archaeological evidence for it. The mechanics of language change itself for either scenario is not convincing. We do have historical examples from the same region where a group of nomadic warriors from the steppes invaded South Central Asia, from the Kushans who are believed to be Tocharian speakers to the various “Iranian Huns” and they all ended up adopting the native languages, and beliefs whether Bactrian or Indo-Aryan. There are also parallel scenarios in Mesopotamia. The Gutians of the Zagros conquered Akkad and then ended up adopting the Akkadian culture, meanwhile, the Shimashkians conquered Elam and ended up as an Elamite empire, and even centuries after the Kassites migrated to Babylon they ended up absorbing the language and culture. Yet these nomads in South Central Asia supposedly had such a linguistic impact from toponyms to theonyms with barely any cultural or genetic input.

Even long ago, before aDNA, when everything that may resemble IE was considered from the steppes I preferred the older Indo-Iranian across the Caucasus model. To me, Early Andronovo was either something related to Thracian or an extinct IE language. Ignoring the general arguments of some stick or stone being Indra’s vajra and red ochre, most attempts at connecting I-Ir rituals with those of Andronovo and the steppes are pretty much grasping at straws, and on closer inspection, some interpretations are just plain wrong, especially those dealing with the Avesta due to the lack of knowledge of the Sassanian and Achaemenid interpolations. Not only that, but a lot of linguists have not caught up to the archaeological findings, let alone aDNA. As an example, in 2019, Michael Witzel wrote the following: “Importantly, even some aspects of typical IA social culture have turned up in the archaeology of the Ural area, such as the Vrātya young men association (Männerbund) that celebrated a ‘dog killing’(śvagn-in) ritual in winter. Remnants of it have recently been discovered at Krasno-Samarskoye, just west of the Urals. In the same area, some 40 years ago, a grave has been found with a headless body; instead a horse head had been substituted, just as in the Ṛgvedic Dadhyañc myth.” How out of the loop can you be? 12 years after this horse nonsense has been disproved by none other than David Anthony (2007): “was from a human skeleton of the Poltavka period that was later cut through and decapitated by a much deeper Potapovka grave pit. A horse sacrifice above the Potapovka grave is dated by sample AA 47802 to about 1900–1800 BCE. Although they were almost a thousand years apart“.

But of course, Anthony also uses Srubnaya’s dog eating as evidence of Indo-Aryans which is surprising. I was perplexed by these dog priests for a while. I searched the RV many times and I found no mention of dog sacrifices or any dog priests. It turns out that vratya in the Rigveda means a group of people. Of course the meaning changed over time and there have been certain periods where it was used positively in the meaning of a wandering ascetic and in others as savages. We know that for Indo-Europeans the wolf is a symbol of murder and dogs were associated with guarding the dead. Indeed, there is evidence of a warband associated with wolves in Europe, but there is simply no evidence of the ceremonial eating of dogs in India, Iran, Avesta or Rigveda, so this cannot be used as evidence for a migration. However, if we want to nitpick there is more than enough evidence in the Rigveda to argue against the whole wolf warband “Verse 3 is quite explicit about the affinity of the arí: he is “one of our own,” while a person who does not belong to the larger Ārya society is called a “wolf,” an outlaw, when his actions threaten us. The poet calls on Soma to destroy both types of enemies”(Jamison and Brereton, 2014). Also, see Hittite Laws (I. 37).

On the other hand, Srubnaya was considered by some in the mainstream as a candidate to be the Proto-Iranian homeland and even the Zoroastrian homeland by Mary Boyce. What we see is the opposite in Zoroastrianism. Herodotus even states that the Magians kill every animal except dogs. Yet somehow Kershaw (2000) found vratyas in the Avesta: “The Devs, of the race of Hesm (Aesma), attack the Iranians, burning and plundering; they wear their hair long and loose, their clothes and weapons are black and they wear a leather belt. These enemies of the lands Ahura Mazda made are thus also his enemies and can be called worshipers of Ahriman.… (The Devs thus look like their worshipers – or vice versa. As we shall see, they also look exactly like the Indian vratyas).”

This text is the Bahman Yasht composed during the early Abbasid era; it contains mentions of Mazdakism, Romans, Arabs, Parthian and Sasanian kings. Most importantly, it is one of the few Zoroastrian sources that describe the rise of the brutal regime of the enigmatic Abu Muslim. In the text, these devils came from Khorasan and destroyed Iran and then the kingship went to Arabs and Romans, or in the Syriac sources: “A man named Abu Muslim began to preach revolt and won many over to his ideas. He had fourteen close followers; they wore black, practised asceticism, let their hair grow long and showed partisanship for the descendants of Muhammad. Many of the people of Khurasan joined them and they became a big party”. So even Abu Muslim was originally a Proto-Aryan vratya wolfman! Also, the color black was not associated with warriors in Indo-Iranian tradition: white was associated with the priests, and red/yellow with warriors.

The circular argument goes as follows: Indo-Iranians are from Andronovo, in Andronovo dogs were eaten thus it confirms Indo-Iranians migrated from Andronovo. There is much more that can challenge the familiar rhetoric of “consensus” and “two hundred years of Indo-European studies led to a steppe homeland.” I am not claiming that the concept of a warband did not exist in Proto-Indo-European or Indo-Iranian times but rather that we cannot reconstruct proto-Indo-Iranian culture to match our narrative using medieval religious polemic. Generally, when it comes to the steppe homeland we are dealing with unfathomable levels of confirmation bias and a tendency to fit the data into models.

 

Dating of Indo-Aryan Attestations and Chariots and Horses

Currently, there is a timeline issue in Indo-European studies. Even after the calibrated dates for South Central Asia were established some scholars ignored or outright became deniers of certain rituals or sites. A surprising example of that is Sarianidi dating Gonur to 1500 to 1200 BCE against his own carbon dating (Lamberg-Karlovsky, 2002). The intent behind that is clearly to conform with the traditional linguistic dating of Zarathushtra. Witzel (2003) has another interesting claim: “we cannot expect Zoroastrian rituals in the BMAC in 2000 BCE but only around 1000 BCE” but why not? This is one of the clearest examples of circular reasoning: “Fire worship is IE but BMAC is not IE so that isn’t fire worship”. Linguists act as if they have Zoroaster’s birth certificate but this is not the first time linguists and philologists have made stuff up when it comes to Zoroaster. Xanthus of Lydia claimed that Zoroaster lived 6000 years before Xerxes but some took it as a mistake stating that it should be 600 years instead, putting Zoroastrianism at their favorite date of 1000 BCE. The funny thing is that the other Greek sources claim that Zoroaster lived 5000 years before the Trojan War. Of course, neither should be taken seriously.

So what’s special about Andornovo? Horses and chariots at a time when the chronology of BMAC and Andronovo was not established. While it is unconfirmed that the Proto-Indo-Europeans used the horse for anything other than food and perhaps milk there is evidence of domesticated horses and carts in the Avesta and Rigveda. Of course, due to the *éḱwos debate, many proponents of the steppe hypothesis deny even the presence of the wild horse south of Kazakhstan and Caucasus. Here, I will list the findings related to horses and carts in South Central Asia for Witzel (2003) claimed: “Most notable is the absence, so far, of horse remains, horse furniture, chariots (invented around 2000 BCE) and clear depictions of horses in stratified BMAC layers. One can hardly imagine the IIr.s without their favorite prestige animal, the horse”.

The truth is that horses were found in Iran and South Central Asia long before BMAC. A wild horse dated to the late neolithic was found in north central Iran along with two dated 3500 BCE and 3000 BCE Godin Tepe (Mashkour 2003). But those in Eastern Iran are more important for the Indo-Iranian debate. Littauer and Crouwel (1979) confirmed horses in South Eastern Iran, “Equus caballus, Remains are reported from recent excavations at Tal-i Iblis in south-central Iran. They belong to two individuals, one from level I (ca 3500 B.C.) the other from level IV (ca 3000 B.C.)”. Bonora also cites Bőkőnyi and Bartosiewicz (2000) for possible evidence of the horse in 3000 BCE Shahr-i Sokhta concluding: “Al momento attuale, la posizione di Meadow non può più essere accettata. Le più recenti ricerche condotte sull’altopiano iranico permettono ora, infatti, di ipotizzare la presenza del cavallo domestico già alla metà del IV millennio a.C”. and on the Surkotada horses: ”M. Tosi (comunicazione personale), che fu al fianco del veterinario ungherese nell’analisi dei reperti, conferma che si trattava di cavallo domestico, ben riconoscibile nella marcata usura dei denti inferiori provocata dall’abitudine, tipica del solo cavallo domestico, di grattare l’apparato dentario contro la mangiatoia in legno, a pasto consumato”. It should be noted that domesticated horses are not necessarily an ethnic indicator of Indo-Europeans, as the Greek and Anatolian version of the Asvamedha it is a donkey that is sacrificed (Watkins, 2004).

As for the chariots it’s been known (Kuzmina 1983) that BMAC proper had horses and chariots, “In southern Turkmenistan during the Namazga VI period, substantial innovations took place in wheeled transport: spoked wheels appeared, and they were used in light vehicles and chariots drawn by horses. The spread of this fundamentally important invention may be dated by finds at Namazga-depe of a model of a wheel with hubs on both sides and with brown paint used to draw in four asymmetrically placed spokes and to outline the hub; at Tekkem-depe, by a model with six spokes painted in in red; and at Elken-depe, by wheels with fingernail impressions possibly also imitating spokes. (75) A clay horse’s head found in the Namazga VI layer in the tower of this settlement (76), and the first certain finds of horse bones in the settlements of Turkmenistan, Kelleli I, Namazga-depe, Tekkem-depe, and Takhirbai 3, also date back to the same time.
The most ancient horse bones in Turkmenia have been identified by N. M. Ermolova at the Kelleli I settlement in the Murgab Delta in a complex dated by I. S. Masimov to the end of the Namazga V period”

And most recently, a horse predating Sintashta has been found in Gonur, “Moreover, it is necessary to add that horse age from the burial N 3200 in Gonur-Depe, which has a radio-carbon dating 2250 BC, i.e. one and half century prior to burial of Sintashta culture dated to 2100-1700 years BC.” (Bonora, 2020). It should be noted that this horse has been buried with a four-wheeled cart with solid wheels. This begs the question (also for the Sanauli chariot): were the four-wheeled carts pulled by horses just like those in Mesopotamia were pulled by other equids or were these just ceremonial and the ones pulled by horses had spoked/crossbar wheels? Even so, while the archaeological records of greater Iran and India are poor in comparison to Eastern Europe there is no denying that the people of BMAC had domesticated horses and chariots.

No surprise that the area with the oldest two-wheeled carts in the world also had access to chariots very early. But all these findings bring me back to the old Littauer and Crouwel model where the chariots developed independently of the steppes from the “proto-chariot” of Hissar IIIB (2500-2200BCE) that had crossbar wheels. Parpola (1995) also confirmed evidence of the horse in the nearby site of Shah Tepe: “Horse bones have been found in Gurgan (Shah Tepe), but not, so far, in Bactria or Margiana. The horse is represented, however, on several of the ceremonial weapons said to have come from the looted graves of Bactria; here one bronze statuette even shows a horse with a naked ityphallic rider, who is pressing his bent legs backwards beneath the horse, without stirrups”. The dates of horse bones in Shah Tepe may be earlier than BMAC, “Of the domestic horse there was only a bone-fragment in layer III, but several bones in layer II”. Another nearby pastoralist site, Tepe Damghani, may also provide early evidence of a horse “a distal tibia was found in upper layers and is more compatible with the horse measurements… The probable presence of horse in the late Namazga IV period is one of the interesting highlights of this study” (Francfort et al, 2008). From Parpola (2015):

The current evidence, even if usually ignored, shows that the carts predate any supposed steppe migration to the region and are a consequence of a long development from northeastern Iran and southwestern Turkmenistan or the region known as Parthia in antiquity which is most likely the source of the Gonur carts.

Even the Indus Valley Civilization had early carts which are now confirmed by the Sanuali burial. From Kenoyer (2004): “During the Harappan Period (Harappa Phase, 2600…1900 BC) there was a dramatic increase in terracotta cart and wheel types at Harappa and other sites throughout the Indus region. The diversity in carts and wheels, including depictions of what may be spoked wheels, during this period of urban expansion and trade may reflect different functional needs, as well as stylistic and cultural preferences. The unique forms and the early appearance of carts in the Indus valley region suggest that they are the result of indigenous technological development and not diffusion from West Asia or Central Asia as proposed by earlier scholars.”

It seems the timeline for the two-wheeled carts in South Central Asia is as follows:

1. 3000-2500 BCE, the development of the first solid two-wheeled carts pulled by bulls in Altyn-Tepe.
2. 2500-2200 BCE, the development of the “proto-chariot” of Hissar and the appearance of DOM2 horses.
3. 2200-2000 BCE, the local development of the spoke wheel or adoption from Sintashta in the greater BMAC area.
4. 2000-1800 BCE, the dispersal of the chariot from the greater BMAC area.

The evidence does not lean toward steppe migrations bringing horses and chariots, at best they were traded. However, it should be noted the current mainstream studies advocate chariot adoption from the steppes but the spread from BMAC. The more likely scenario is that these domesticated BMAC horses come from northeastern Iran which probably acquired them from Transcaucasia ultimately going back to a Yamnaya-related source shortly after the domestication of DOM2 horses. Indeed, Anthony (2007) proposed a similar scenario for the early horses in Iran: “A few horses might have passed through the Caucasus into western Iran before 3000 BCE, indicated by a few probable horse teeth at the site of Qabrestan, west of Teheran” but those are too early to be the DOM2 horses. There is also much more evidence for horses post 2000 BCE but that is expected.

 

Literary sources

The Mesopotamian literary sources contain valuable information about horses. For example, there is considerable evidence that horses were being ridden in the 18th century BCE as evidenced by the Tell Leilan texts. In a letter (L.87-651) from Sepallu to his brother Mutija where he rode a horse with 60 men and went to battle. It is unknown if he rode the horse for a quick raid or if this is early evidence of cavalry (unlikely). Other texts also mention that horse riding was the way of Amorite kings, not chariots. Yet even earlier than that during King Shulgi’s time horse riding was not considered for royalty but donkey riding was.

But what matters is where these horses are from. Our first clue is that the sisi (from Indo-Aryan?) or ANŠE.KUR.RA, which means ass of the mountain. Both Sumerian and Old Babylonian texts confirm that these horses were brought from the western Zagros (Lullu and Simirrum), and even Elamites were involved with horses and Anshan was quite advanced since has the earliest equid guided by a metal bit in the world (Potts, 2014). However, there is an early mention of the mountain sisi in the Lament for Kesh (2400-2300 BCE) that caused some to discard the idea that sisi is a loan from Aryan.

 

Linguistics

There is little agreement between linguists that support the same homeland on the basics of PIE, let alone if they knew of lions, leopards or elephants or not and what kind of agriculture they practiced. For example, according to Lubotsky (1999) on the way south the Indo-Iranians borrowed some words from BMAC, some are general like smell and hair but others were interesting like camel (ushtra). Then again, Mallory (1997) argues that ushtra could be derived from *usr (auroch). Lubotsky even considers Indra, who has a clear IE function, to be loaned from BMAC, but why? Well, because Indra is attested in Mesopotamia by the end of the third millennium BCE as such he cannot fit within Lubotsky’s time frame. Let’s ignore the proposed Indo-European etymologies for these supposed loanwords. Why not consider BMAC Indo-Iranian and let’s say these loanwords come from Shahr-Sukhteh, Mundigak, Sarazm, or vice versa, or why not from somewhere east of Elam? Indo-Iranian having a substrate or adstrate is not unlikely considering the geographical location.

Not only that, but linguists have somehow found relationships between PIE and whatever languages are spoken in the neighborhood of their preferred homeland, from Caucasian to Uralic to Semitic. In general, while I think we have to avoid a strong stance on such subjects, there is one model that caught my attention recently: the Indo-Burushashki model of Eric P. Hamp specifically where he goes against the grain and considers Proto-Iranian to have split before Proto-Indo-Aryan. Such implications will be considered when looking at the Rigveda and Avesta.

The loanword arguments do not support any homeland. Many linguists argue that the oldest Indo-Iranian loanwords are found in the languages of the North-Eastern Caucasus with Harmatta pointing out that the earliest traces of contact are found in the Udi language of Caucasian Albania. Proponents of the steppe homeland use it as evidence of a west-to-east movement where Indo-Iranians are in contact with North Caucasians first and Uralics later. The funny thing is that Ivanov and Gamkrelidze also used that same argument as evidence for their Armenian homeland. Also, while there is an agreement that the direction of the loanwords is from Indo-Iranian to Uralic there is no consensus on where the Proto-Uralic homeland is located nor which Indo-Iranian language is the origin for these loanwords that is for Lubotsky considers them to be from Sanskrit, Diakonov Indo-Iranian, Gamkrelidze and Ivanov Iranian, and Joki Middle Iranian.

While some supporters of the steppe hypothesis group Indo-Iranian with Greco-Armenian, others argue against such a group. Despite Iranians interacting with different groups across the world from Hungarian to Japanese, some modern day Slavists argue for a Proto-Indo-Iranian or Indo-Aryan relationship with Balto-Slavic and try to downplay every supposed relationship with Iranian. Let us give in and accept that Slavic-Iranian contacts were limited, which would be unlikely considering that the Iranian Scythians were very close to the proposed Balto-Slavic homeland. The real problem with this hypothesis is the assumption that all the Scythian groups were Iranian. In Scythia, the Greeks mention an agricultural tribe called the Sindoi now believed to be an Indo-Aryan speaking group, similar to the Mitanni.

Witzel assumes that the Indo-Aryans borrowed Kubha and Sindhu from North Caucasians and brought them to South Central Asia. These Sindoi are thus considered the original Indo-Aryans who stayed behind. Yet elsewhere Witzel notes the Mesopotamian texts mentioning the Sinda (Sindhu) wood and notes that Sindhu could be from Burushaski. Despite this, the Greek sources tell a different tale (more on that later) that goes against the ridiculous underlying idea that South Central Asians were somehow stuck. Not only did South Central Asian religions make it to East Asia but even Northern Pakistani Indo-Aryan languages like the Niya Prakrit were carried by Central Asian nomads. Also, good luck to whoever is looking for a migration from the MLBA North Caucasus to India.

Interestingly, out of the thirty seven Vedic rivers only the Kubha, Sutudri, Ganges, and maybe Sindhu do not have a clear IE etymology (Witzel 1999). One of the problems with a late Indo-Aryan arrival to South Central Asia that is often repeated is the hydronyms since river names are usually archaic and cannot be loaned. Almost all of the hydronyms were Indo-Aryan from the Punjab to Afghanistan and Baluchistan from the Oxus to the Gomal already by the bronze age Rigvedic era. In the Younger Avesta only two toponyms are considered to be non-Indo-Iranian by Witzel and non-IE names are absent. In comparison with the Iliad which was composed almost a thousand years after the arrival of the Greeks, allowing a longer time frame for them to impose their language, only one third of the toponyms are Greek.

 

Near Eastern Indo-Aryans

Most important is Indo-Aryan attestation in the Near East. The language of the elites of the Mitanni empire when first discovered was considered an Old Indo-Aryan language with some Iranian elements. But now it is considered to be just an Old Indo-Aryan language, maybe even proto-Indo-Aryan with Witzel claiming that it is older than the language of the Rigveda based on sound changes. However, dating oral poetry based on simple sound changes is not that convincing, “The fact that proto-Aryan *ai and *au are replaced in Indo-Aryan by e and o, while in Iranian they are preserved as ai and au and that ai and au regularly appear on the Anatolian documents (eg. Kikkuli‘s aika), is unfortunately inconclusive. It is quite possible that at the time of our oldest records (the hymns of the Rigveda) the actual pronunciation of the sounds developed for *ai and *au spoken and written by the tradition as e and o, was still ai and au. The e and o can be a secondarily introduced change under the influence of the spoken language or the scholastic recitation” (Thieme, 1960). But OIT supporters have also latched on to the sapta (Rigvedic) > satta (Middle Indo-Aryan) to prove that the language of the Rigveda is older than that of the Mitanni. Witzel (2001) argues that this satta of the Mitanni is due to Hurrian writing systems and that it is influenced by the Hurrian number seven, šinti but also notes: “Some of the so-called MIA features of Mitanni-Indo-Aryan are due to the writing system (in-da-ra, etc.); satta is questionable as well: ša-at-ta is influenced by the Mitanni term, as *ša-ap-ta would be possible in this writing system”. However, this sapta > satta change occurs even in the language of the Sindoi who are supposed to be the leftover Proto-Indo-Aryans so is it also due to Hurrian cuneiform influence?

As to when the Aryans reached the Near East, Harmatta proposes that the Indo-Aryans were behind the Gutian conquest of Mesopotamia and that Indo-Aryans ruled the Hurrians by 2300 BCE, but the supposed Aryan names in 2300BCE amongst the Hurrians are no longer accepted. He also notes Proto-Indo-Iranian influence, “Assyrian sources preserved the Lullubean word kiurum ‘god’ which can be regarded as an adoption of PII *kura- (cf. OInd sura-, Avestan süra-), the Old Indian and Old Iranian correspondents of which were still applied to denote some gods in Vedic and Avestan times”. He proposed that the Near Eastern Aryans came from Northeastern Iran, which is generally accepted but under varying dates (1500 BCE for Young, 1760 BCE for Parpola, 3000 BCE for Ghirshman) and confirmed by the toponyms in northern Iran (eg. Mount Bikni from Indo-Aryan Vighna and Lake Urmia). Roiter (2013), also proposes that the Gutians (2100-2000 BCE) had Indo-Aryan elite, “The name Indar relates to the Gutian theophoric Inda-ššu, similar to the Kassite Indaš and the Sanskrit term indrasūnu “son of Indra”. Indaššu was the name of an Ensi of Zabshali recorded in a tablet dated to the Ur III period”. If Indassu means “son of Indra” then we have a prototype of the Mitanni title of “son of the storm god” who in that context is Teshub.

Naturally, Witzel argues against Harmatta’s model but does not deny the Indo-Iranian influence in the Near East evidenced by the spread of BMAC related material: “It is probable that this move was preceded by successive spearheading forays of (non-IIr. speaking) mountain peoples into Mesopotamia, such as the Guti, Lullubi, and Kassites (c. 2250-1750 BCE), who were as yet only marginally influenced by IIr. languages and customs”.

While Witzel proposes a 1600 BCE entrance for the Mitanni, Harmatta was not the only one who opted for a third millennium entrance. Earlier Kammenhubber also insisted on a third millennium Hurrian-Aryan symbiosis but that is no longer accepted. The most reasonable date came from Wilhelm (1989), “Aryans had made their appearance in Northern Mesopotamia along with the second wave of Hurrians in the 18th-17th centuries B.C. Aryans together with the new groups of Hurrians must have descended from the southern parts of the Armenian Highland, infiltrated the future territory of Mitanni, and had subsequently played an important role in the formation of that state” (Kosyan, 2006). Not only does this match with the letter of Tell Leilan (L.87–887) found later, where it mentioned maryannu warriors being employed around 1760 BCE but also matches with other texts mentioning elite soldiers coming from the direction of Iran from the same period. “Finally mention must be made of a third kind of international contact, that relating to individuals or group of foreign extraction mentioned in the texts as present in the région. All references of this order are to people from the east, from Kakmum, Simurrum, Simaski and Gutium, and reflect a phenomenon attested also in the older northern corpora of texts from Mari and Bimah, where people from the east-Tigris or Zagros régions seem to have been widely used as mercenaries and employed also as a kind of “Swiss” guards by several kings” (Eidem, 1987). There are currently a few variations of this generally accepted 1800 BCE entrance of the Mitanni but an interesting one is Van Koppen’s (2004), where a group of mercenaries who derive from deportees during Hammurabi’s wars in the Zagros. Van Koppen’s (2017) most recent hypothesis, which I disagree with but is worth mentioning, is that the migrating Kassites brought the Aryan elements along.

There was likely at the very least one Indo-Aryan group active in the Zagros. Jesper Eidem (2014) hints at the most likely scenario that the rise of the Mitanni royal family is related to the wars in the Zagros during the time of Shamsi-Adad. The powerful Gutian king Indassu warred with the Assyrians, Esnunnaeans, Turrukkeans, and Elamites. This war led to the migration of Itabalhum Hurrians from the Urmia Basin, an Indo-Aryan hydronym (urmi- “wave”) where BMAC related material is found, to the west. I suggest that the Indo-Aryans involved in this war, who originally came from northern Iran, influenced Gutian, Shimashki, and Kassite groups. The family of Maitta were the direct descendants of these Indo-Aryans as they are the only group that chose Indo-Aryan throne names even after migrating with the Hurrians and adopting Hurrian names and identities.

The real takeover from all of this is that by 1800 BCE there was a group of Indo-Aryans that settled in western Iran. Also, it should be noted that maryannu was not a technical term as is suggested in many other blogs: “In most parts of the Mittani Empire, the nobility were dignified with a special designation, maryanni, which was formed by adding the Hurrian derivational suffix -nni to the borrowed Indo-Aryan word márya, “(young) man”. In Arraphe, the same class was instead denoted by the Akkadian designation rākib narkabti, “chariot rider”. It bears emphasizing that the term maryanni has nothing to do with chariotry; etymologically, the word simply means “man”, although as a modified loanword it was semantically transparent to no one. As such, maryanni was used to denote a special class of man, the nobility. Because the most distinctive feature of this class was the use of chariots in war, the term for it eventually acquired the connotation “chariot warrior”, as its use spread beyond Mittani” (von Dassow 2014).

Lastly, in the third millennium, the Near Eastern texts mention a land to the east of Gutium rich in lapis lazuli and gold called Tukrish, its people along with the Guti were called by W. B. Henning “The First Indo-Europeans in History”. He considered them to be Tokharian speaking, but the historical Tokharians spoke Bactrian and even their ethnonym is Iranian. “The word Tukriš has been compared with Tuγrån, Tuγrastån, Tokharoi, etc. (Henning), see EWA I 651 s.v. tugra… Only some initial guesses are possible, for example about the ethnic nature of the Tukriš which might be connected with Ved. tugra, tugrya (both personal names), Iran. tuγr-.” (Witzel, 2002). Even though Turan/Tugran is usually associated with East Iranians, Tugra is also a Rigvedic hero.

The earliest recorded Indo-Aryan and Indo-European deity is Indra in Ur III (Blazek, 2003 & Roiter, 2013). From the same period, Blazek (2003) also considers the minor “Elamite” deity Aguni to be Agni. Blazek also considers some of the names of the Vedic demons to be Elamite. I find some of these Elamite names to be incredibly unlikely, such as Susna being Inshushinak, the Tutelary god of Susa, who probably was not a major deity beyond Susa.

Assuming the Mitanni empire began in the 16th century, a list of pre-empire Indo-Aryan related names in the late third millennium, Ur III (2200-2000BCE):

• Indar, a minor god in the Gudea inscription and also in the Mitanni treaties
• Intari and Inda, unknown context (related to Kassite Inda and in that case Indra but Zadok considers this to be explicit in Elamite terms)
• Indassu, governor of Zabshali in the Zagros kingdom of Shimashki

In the early 2nd millennium:

• Indassu, king of Gutium during the time of Shamshi-Adad (1808-1776BCE)
• Agni along with Hurrian leaders with IA names Karawani, Parayuna, and Ayuktaeraya in CTH 13 (1650BCE)

The linguistic evidence is very limited considering that some of the Zagros soldiers adopted new names under their Mesopotamian employers as is the case with the Gutians who mostly had Akkadian names. In the Girsu texts one of the “men of the mace” adopted the name Šulgi-Aguni who appears in the context of soldiers from Susa, Sabum, Anšan and Šimaški, i.e. the east.

While the movement of Indo-Aryan elements is clearly from east to west in terms of the late third millennium it is difficult to claim that there was an Indo-Aryan movement to southern Mesopotamia. To be cautious, for the third millennium, I will only consider Indar the deity to be actually Aryan. Most of the early names do not have a clear context as theoretically Indra could have been honored as a minor deity of powerful trading partners as is the case with the Elamite Inshushinak. IVC and BMAC contacts with Mesopotamia are well established as BMAC adopted the kaunakes dress and the Oxus goddess from Mesopotamia and the “Zimri-Lim Palace” was built by Gonur craftsmen.

 

Parameters For the Indo-Aryans

I propose a few basic parameters before getting to archaeology and the Indo-Iranian religious texts. Any hypothesis for the Indo-Iranian homeland has to explain the following:

1. Long distance Indo-Iranian contacts with the groups around Mesopotamia in the late third millennium BCE
2. 1760 BCE Indo-Aryan maryannu in Mesopotamia
3. The complete Indo-Aryanization of the hydronyms in the entirety of South Central Asia when the Rigveda was composed
4. The Rigveda being composed in the bronze age
5. Indo-Aryan presence in North Iran (Mount Bikni and Urmia)
6. Indo-Aryans north of the Caucasus

 

The Indo-Iranian Religious Texts

The Avesta is a very tricky book to deal with since someone who does not understand the context of Iranian propaganda can fall into some weird rabbit holes of interpreting the rise of Abu Muslim as evidence of Proto-Indo-Iranian werewolves. The supposed author of the Avesta, Zarathushtra, only composed the Gathas and the Yasna Haptanghaiti most likely composed by his followers. These two small corpora of texts are what is considered the Old Avesta, which unfortunately contains very little information on their time and place. The Younger Avesta is composed in a different language with various additions over time and even some of its hymns are considered an incoherent mess. Some of these Younger Avestan texts are considered to have been composed in the Parthian and Sassanid eras. Many of those who proposed a Medo-Persian timeframe for Zoroaster fall into the trap of considering these texts as from his time. The Avesta itself doesn’t even mention one of the attested great Medes and Persians other than a Vishtaspa who doesn’t have a son named Darius. The other argument of the language Gathas is a result of the priestly training that taught Zarathushtra an ancient language doesn’t hold any water since it doesn’t explain why the other priests could not replicate the language of the Gathas. In my opinion, there are three main traditions in the Avesta, Older Avestan, Younger Avestan, and Western Iranian.

As for the Rigveda, it is a book composed for the elite by the elites, ie. those who know. Most of the stories and descriptions in the Rigveda are vague and metaphorical because the intended audience already knows them. Unfortunately, that is not taken into account since usually when studying the Rigveda the absence of evidence is taken as evidence of absence. It is as if the Rigveda is some archaeological list or catalog. However, at other times certain pieces of evidence are acknowledged but under a different meaning, eg. pur (fort/city) as mud rampart. But let’s save that for later and get to the Avesta first.

 

Younger Avestan Homeland

The most well known book that mentions the Avestan geography is the Western Iranian book the Vendidad. Despite its late composition, which is considered to be during the Parthian or Sasanian period due to its corrupted Avestan, the book contains traces of some very old myths so it cannot be ignored. For its date, I lean more on the Sasanian side since the book might contain the Aswaran equipment list of Khosrow’s reform (Vdd. 14.9-14). In my opinion, it could very well be related to the Kushano-Sasanians as well. There are many disagreements on the sixteen perfect lands of the Vendidad and I think it is mostly because it is considered a genuine Avestan era text and the Middle Persian texts are not being taken into account.

Many have identified the first of the sixteen lands, the cold Airyanem Vaeja with the oasis of Chorasmia due to the idea that the Iranians came from the north, ignoring the fact that the north was associated with Adurbaygan in the Sasanian era. Why is Adurbaygan associated with Airyanem Vaejah? The most logical answer is that after the fall of the Achaemenid Empire, Greek influence spread across the Iranian plateau but the land of Atropatene held on to its Iranian culture and became the most important place in Zoroastrianism and where its most glorious temple was built and the swift Daitya river that is full of snakes is none other than the Araxes that is also full of snakes. Even the ordering of the lands doesn’t make much sense, “If we would try to list these countries consecutively, along a strict counter-clockwise path some disruptions appear…This list thus “does not work” in the usual IIr. fashion” (Witzel, 1998).

If there ever was an Airyanem Vaeja during the time of Zarathushtra its location has long been forgotten by those who composed the Vendidad. The second land is Gava, the homeland of the Sogdians (Persian form) is also evidence for the late composition of this text. What is interesting is how the eleventh land, the Helmand that flows into the lake of the Saoshyant, is now a place of witchcraft indicating a foreign religion.

The oldest layer of the Younger Avesta, the Zamyad Yasht describes a southern homeland around the Sistani lakes. Both Burrow and Gnoli support this Sistani homeland and Burrow dates this text to five generations after Zarathushtra and before the migration of Iranians to Media. In my opinion, religious genealogies should not be accepted without being analyzed even though I do agree that it is very likely that if Sistan was not the homeland of Vishtaspa and the Kayanians then it would have been one of the early centers of Zoroastrianism. It should be noted that the Kayanian genealogy might be fabricated, e.g. the famous Kayanian Kavi Us who is also mentioned in the Rigveda (Kavya Usanas) who might be a Proto-Indo-Iranian sorcerer according to Dumezil. In the Mihr Yasht, we are introduced to Airyo Shayana which consists of the lands of Ishkata and Pouruta, Mouru and Haroyu, Gava-Sughdha and Xvairizem. Ishkata and Pouruta are mountains mentioned in the Zamyad Yasht identified with mountains near Herat and the Helmand, the latter three are well known from history Margiana, Aria, Sogdiana, and Chorasmia. However, Sogda and Xvairizem are believed to be in Western Iranian forms meaning they were added relatively very late.

Zarathustra’s homeland itself is a mystery. Humbach (1991) believes that Mozduran, basically on the eastern border of the later province of Parthia, is an Old Avestan toponym which if true makes it the only known one. Parthia is a region that is heavily associated with Vishtaspa and Zoroaster in the later traditions as Kashmar is one of the places where Vishtaspa converted. Based on the Younger Avesta, a homeland in Greater Parthia is as likely as one in Drangiana (Helmand, Lake Kasava, Mount Ushidhau) and Aria (Ishkata and Vaiti-gaesa, i.e. Badghis). Of known Parthian mountains, Mount Raevant of the Avesta is Mount Revand in Nishapur where the later temple Adur Burzen-Mihr was built, and Mount Spento-data another Mazdian one.

This Yasht homeland matches very well with Strabo’s Ariana which is bounded by the Indus in the east and the Paropamisadae (Hindu Kush) in the north and extends to Carmania and the Caspian gates in the west. He also mentions that the name of Ariana later extended to Persia, Media, Bactria, and Sogdiana. On the other hand, Pliny mentions that some geographers consider the eastern parts of this Ariana to be Indian which seems to be more accurate as Arachosia was called white India. Of course, there are other tribes in the Avesta that could have lived outside these lands, the Turyas, Sairimas, Sainu, and Dahi. Historically, the Dahi have been connected with the Murghab or the Ochus, and Turanians either with Khuzdar or Transoxiana.

 

Indo-Aryans of the Avesta

The Gathas describe a world that turned to violence, the main antagonists being Daeva worshipping priests and warlords. Surprisingly, there is no mention of Aryans in the Gathas and the children of a certain Tura are honored. Also, there is no mention of Arjaspa the main rival of Vishtaspa nor his Hyaonas. According to Burrows (1973), the original Daeva worship of the Gathas comes from Indo-Aryans as the Iranian word for god was baga/yazata. Zarathushtra’s Wise Ahura shares functions with the Asuras of the Rigveda Mitra and Varuna as both are wise and in charge of order/truth that his original name is considered by some to be Asura Varuna or Asura Mitra. That would explain why Mithra was not explicitly condemned in the Gathas and it would mean that Zoroastrians readopted Mithra later as a separate entity. However, since Varuna has a similar counterpart in Apam Napat, and the Yashts give us another clue as it seems Mithra fell out of favor with the Zoroastrians and was later reintroduced during the Younger Avestan period (Yt.10.54-55) the most likely candidate for who Ahura Mazda was based on is none other than the original Indo-European and Indo-Iranian skyfather, the Asura father Dyaus (Scythian Papaios).

Either way the world reflected in the Gathas is purely an Indo-Iranian one. The Gathas unfortunately do not describe the homeland of Zarathushtra but there are a few interesting facts. The Iranians of the Gathas had domesticated horses and camels. We have evidence of chariots as Zarathushtra metaphorically wishes to spread his message as a charioteer (Y30). It is interesting to note that unlike in the Rigveda, the chariot terminology is poor in the Avesta in general, where Ratha meant the chariot box and there is no mention of words for axle or parts of the wheel. Adding to that in the Younger Avesta the warriors are called “rathaeshtar”, ie. charioteers but in the Gathas, they were called “nar”, man and the enemy warriors in the Younger Avesta mairyo ie. marya of the Rigveda and Mitanni.

In many passages of the Younger Avesta, the heroes of Iran pray to the gods so that they can defeat their enemies. The Turanian Afrasiab and the demon Azi Dahaka are well known but there is another interesting group, the daevas of Mazena and Varena. Mazena is Mazandaran but Varena is disputed. In the Vendidad, we get the names of the daevas of which three are the Indo-Aryan gods Indra Sarva (Rudra) and Nasatya (Ashvins). These devils are none other than the western Indo-Aryans, the Amardians who went on to rule the Mitanni empire. Indeed, even in later myths, Rustam faces a demon called Kulad Ghandi in his conquest of Mazandaran. There is also evidence that the Indo-Aryans have survived in the west in pockets till the Median era as there is a certain Agnuparnu mentioned in a letter to Ashurbanipal. Of course, not all Indo-Aryans were considered enemies since the Farvardin Yasht honors a son of Dashtagni who is from Muzhavant and is probably a recent convert.

 

Timeline for the Younger Avesta

While omitting the Old Avesta we can at the very least establish one for the Yashts. The soma is brought from Mujavant in the tenth book of the Rigveda, yet by the times of the Atharvaveda, Mujavant is considered to be a foreign country. It is located next to Gandhara and Bahlika (Balkh), which means its location was in the eastern Hindu Kush thus Muzha of the Farvardin Yasht is the same as the Mujavant of the Vedas, which is long gone by the time of the Achaemenids. Dashtagni is an Indo-Aryan name so the conversion of the region was happening during the time of the Yasht. Thus the Atharvaveda was composed after Mujavant became Iranian or Zoroastrian. Not only that but the Atharvaveda contains the first mention of Bahlika which is missing from the Yashts and only mentioned in the Sasanian Vendidad. So the Farvardin Yasht, which probably was composed over a few hundred years due to its various additions, was composed after the Rigveda but before the Atharvaveda. The first text that mentions iron weapons and distinguishes them from bronze is the Atharvaveda and the earliest evidence of iron weapons in Uttar Pradesh, where the Atharvaveda was composed, is in two iron arrowheads in Ahar dated to 1275 BCE. Then we have to assume that the conversion of Mujavant happened in 1275 BCE at the latest.

From the Yashts, it is clear that there is no evidence for a northern homeland and that if Iranians came from the north they have stayed in Drangiana for quite some time to completely forget it before the West Iranians migrated, which is the position of Burrow. Also, it seems that some of the Younger Avestan texts were revised or changed by Western Iranians, indeed the concept of the Bactrian Avesta is not supported by the Aramaic documents as they provide evidence that the most important deity of Bactria is Vaxshu (from Indo-Aryan vaksu?) who is a non-Avestan god.

However, even if some of the Yashts were originally composed earlier than 1200 BCE the language is still different from that of the Gathas as is the original monotheistic or henotheistic message of Zarathushtra which evolved into polytheism by those times. Saena the Sistani teacher mentioned in the Farvardin Yasht is believed to have lived 100 years after Zarathushtra based on the traditions but that cannot be since there is also a loss of information even though the Yasht is supposed to be a catalog of the Zoroastrian community yet it does not contain any mention of who succeeded Kavi Vishtaspa. In the Persian tradition, Vishtaspa was succeeded by Kay Bahman who is not even mentioned in the Yashts and was killed by a dragon or defeated in northern Iran, and then the kingship was lost. There must have been a significant time, enough to change a priestly and perhaps dead or dying language, between Zarathushtra and Saena who most likely restored the teachings of Zarathushtra, in an abridged pan-Iranian form. In this case, the spread of Zoroastrianism should not be associated with the conquests of Vishtaspa and Spentodata but rather gradually with a community that came after them.

 

The Gods of the Rigveda

At first glance, the most important deity is without a doubt the Deva thunder god, Indra. His association with birds of prey and dragonslaying is pretty typical of Indo-European thunder gods. While it is a controversial topic, I am on the side that considers the sky father and the thunder god to be originally the same deity. Nonetheless, he is missing an important function of the king of the gods found in other Indo-European cultures.

In most Indo-European religions the king of the gods is both conqueror and lawgiver, king in war and peace. Even though in a few hymns he is described as the “lord of settlements” Indra is mostly associated with war, he is a destroyer of forts. For that reason, many wrongfully believed that the Indo-Aryans were nomadic conquerors. The ethical king in peace, Varuna, is not missing in the Rigveda and is as important as Indra. Both Varuna and his brother Mitra have Iranian counterparts in Apam Napat and Mithra. The Iranian deity Verethragna, who is Mithra’s sidekick, may be seen as a counterpart to Indra but he is mostly a god of victory and lacks most of the weather and dragonslaying functions of Indra. The older theory was that Indra was demonized by Zarathushtra because of his violent acts. The same can be said of Rudra who is described as wrathful and terrifying. However, the argument cannot be true for the Ashvins who are protectors and healers. Though it has to be noted, that even Iranian gods fell out of favor after the Gathas and had to be reintroduced during the Younger Avestan period (Yt.8.23-25, Yt.10.54-55).

The currently accepted theory is that these condemned gods were never part of the Iranian pantheon and were only worshipped by Indo-Aryans. Indra is believed to derive from an older Indo-Iranian god, *Vrtraghan. But more importantly, we have confirmation from the Rigveda (X.124) that tells us that the “Asuric deities” that are also found in the Younger Avesta Agni (Atar), Soma (Haoma) and Varuna (Apam Napat) switched sides from the Father Asura to Indra, this father Asura none other than the sky father Dyaus Pita. “On putting these clues together, it seems possible that the old sky-centered religion focused on the inherited paternal divinity Dyaus quietly gave way to one located in realms closer to men” (Jamison and Brereton 2014)

This confirms that the original Indo-Aryan god was derived from *Dyeus who then was taken over by the newer god Indra sometime before the composition of the Rigveda. Indra is also found in the Nuristani pantheon as Indr the thunder god but the dragonslayer is the king of the gods Imra (Yama) (Witzel, 2004). He could have been borrowed from the Indo-Aryans or he could have been an Indo-Nuristani god if such a group ever existed as Hamp proposes, that gained prominence during the Vedic era.

It seems the Aryans who composed the Rigveda were far from forgetful, and the older Indo-Aryan gods were the ethical asha/rta based ones and the warlike Deva ones gained prominence later. Most importantly, the Aryans who ruled in west Asia according to Parpola, Ghirshman and Harmatta came from northeast Iran, followed a religion similar to that of the Rigveda as evidenced by the attested gods in their treaties. Even more so the ordering of the gods in the Mitanni treaties is the same as that in the supposedly late book 10 (125).

I bear both Mitra and Varuṇa, I Indra and Agni, I both the Aśvins

While Agni was missing from the treaty the ordering is the same. Mitra-Varuna for the priest function, Indra for the warriors, and Nasatya for the commoners. For the Ahura Iranian version of the, it would be; Ahura Mazda, Mithra, Apam Napat.

 

Vedic Homeland

Currently, there are two theories for the homeland of the early Rigveda, one in Afghanistan and the other in Punjab with most kurganists favoring the first. The location is determined by the description of the Sarasvati that is identified with the Helmand/Arghandab or Ghaggar-Hakra. But also, some hymns of the RV like in book 5 (54.9) contain mostly mentions of lands west of the Indus and the description of the Sarasvati could indeed fit with the Helmand. As for the samudra meaning sea, even in the Middle Persian Bundahishn, the Kayansih Lake of Sistan is called a sea. On the other hand, there is a consensus that the later books were composed in the eastern area of the Indus.

However, some parts of these later books could have been contemporary with that of the earlier books and the families who composed the RV could have had different Sarasvatis considering the maximum possible timespan for the composition of the Rigveda is 900 years (Witzel, 2003) we better count interpolations, redactions and fabrications.

If we are to ignore the Helmand vs Ghaggar-Hakra argument and come up with a larger homeland that encompasses both the Early and Late Rigveda we will get a much better picture. The westernmost river in the Rigveda is the far Sarayu River, most likely the modern day Hari River, where two Aryans were killed by Indra. The easternmost river is the Ganges, whose appearance is considered suspicious by Witzel. If it is the Oxus the northernmost river is the Rasa (Av. Rangha), which is also one of the boundaries in the Younger Avesta if not then the northernmost location is Mount Mujavant, where the Soma comes from, which appears in book 10. There is no hint of anything north of the Oxus, the Aryans were completely unfamiliar with that area nor is there evidence of a previous homeland for the Aryans, wherever the Aryans came from previously they lived centuries in South Central Asia before they composed the Rigveda.

 

Dasa

The Dasas are the second largest group that appears in the Rigveda but are missing from the Avesta. They are described to be living in the fortresses of the mountains. Over the years many theories have tried to explain who the Dasas are, from mythological enemies to non-Aryan natives to pre-Rigvedic Aryans. Witzel believes these Dasas are ancient indigenous people who were conquered by the Aryans and became mythicized over time. But surely going from human to demon must have taken a while. Though there is one Dasa that appears in the late Rigveda in the “Iranian” book 8, named Taruksa Balbutha, argued to be a non-IE name, who may have been one foot in the Vedic culture appearing in a context where he gives the poet camels.

Of course, some of them have Aryan names which leads to Parpola’s theory. Parpola proposes that the Dasas were related to the Indo-Aryans and the first wave of Indo-Iranians who took over the BMAC very early and spread over South Central Asia. He notes there is no evidence of conquest and that the takeover must have been peaceful. The Aryans of the Rigveda thus represent the second wave of warlike Aryans who conquered the Dasas and later made it to South Asia. The Dasas being pre-Vedic Indo-Aryans can explain why they were quick to adopt Aryan (Rigvedic) culture.

There can be no doubt that the mountains of South Central Asia have been a haven for many smaller groups, IE and non-IE from Burusho to Nuristani, (even before that see the Goat Cult of Darra-e Kur). Yet, looking at the Dasa-Arya conflict I cannot help but question most of the current conclusions. Most of the Dasas in the Rigveda i.e. the Danava sons of the river Danu and enemies of Indra (Srbinda, Susna…etc) cannot be historical; they are mostly described as dragons or demons; they live in the mountain forts made of metal but none of the BMAC fortresses are mountain fortresses. Archeologists do not find evidence of steppe conquests in the second millennium either, currently, the only evidence of burnt cities and settlements we have is in the 4th and 3rd millennium (Hissar, Anua culture, SiS…etc), and if they were in India in 2600BCE Kot Diji phase.

This seems to be a common Indo-Iranian topos also found in Iranian myths mainly in the Azhi Dahaka, Arjasp, and Afrasiab myths where they are associated with metal forts. Of particular interest is Azhi Dahaka’s Sambaran fort which sounds similar to Sambara, the enemy of Indra, which matches perfectly with Parpola’s (2015) reconstruction: “In the neuter gender, śambara- is used to mean “wall, rampart, fortification.” In this sense Sanskrit śambara-/saṃ-vara- and (saṃ-)varaṇa- correspond to Avestan vara-, “fortress”. The only logical conclusion is that these mountain fortresses are nothing but an updated Indo-Iranian version of the original IE dragonslaying myth, especially since in most of the stories related to them either Indra releases the cows or the waters or the Iranian heroes rescue the princesses and avenge a cow.

To simplify it, similar to how the Sasanians used Turanian for East Iranians and Turks later, “Dasa” is an umbrella term for foreigners, those who are not a part of the Vedic culture or speak Indo-Aryan and Dasyu is the follower of non-Vedic religions. The Dasas can be categorized as follows: the demonic enemies of Indra (Vrtra..etc), Indo-Iranians who opposed the Indra cult (the Iranian Arsasana and Varcin), and non-Indo-Iranians (Taruksa Balbutha). There also seems to be evidence of infighting between the Arya Indra worshippers. Ironically, the Younger Avestan Zoroastrians called the Indo-Aryans “daevas” and the Indo-Aryans called some Iranians “dasas” when both groups, as far as we know, identified as Aryas.

There is another group related to the Dasas called the Panis, who are described to live beyond the river Rasa in a Vala and are rich in horses and cattle (X.108.7). The Indo-Aryans are not familiar with this region and it is almost mythical. These Panis are none other than the East Iranian Parni tribe. According to Witzel, they lived where the historical Parni did around the Oxus in Eastern Bactria and were encountered by the Aryans on the way down. Parpola assumes they were Iranian and Witzel assumes they were originally non-IE Dasas who became Iranians later. According to the steppe theory, Aryans came from the IAMC they must have stayed around the Oxus. Yet the Panis were not Indo-Aryans even in the late Rigveda (X.108.8). Funny enough, this region beyond the Rasa is where the steppe impact is largest. This is another clue of the direction taken by the Indo-Aryans and it was not from the IAMC to Swat.

There is another more likely interpretation based on book 8 (66.10) where Indra supposedly dominates the Bekanatas and the Panis. Here Bekanata means usurer and Pani miser, especially in the sense of trader, as put forth by Griffith long ago: “pani (literally, one who barters and traffics) means a miser, a niggard; an impious man who gives little or nothing to the gods. The word is used also as the name of a class of envious demons watching over treasures”. Now, if they represent a merchant class then it can be seen as a religious conflict between the priest and warrior elites who are represented in the Rigveda and some merchant groups, especially if their association with the Ganges is taken into account (6.46.31). While I am tempted to say that the most likely location where this Arya-Pani conflict happened was around the Indus it could also be that the “Pani” referred to both Iranians and merchants so it is inconclusive.

 

Iranians of the Rigveda

Finally, we come to the Iranians, the most important foreign group in the Rigveda who at times are sponsors of the Vedic poets and others as enemies. The biggest evidence of them comes in book 8 (Witzel 1999): “Book 8 has long been connected with Eastern Iran: K. Hoffmann (1940 = 1975: 1 sqq.) has pointed to Iranian looking names such as Kaśu ~ Avest. Kasu- (EWA I 330), Kaśu Caidya 8.5.37, Kanīta ~ Scythian Kanitẽs, cf. further Tirindira 8.6.46 ~ Tiridatẽs ~ Avest. Tīrō.nakaθßa, Krśa 8.59.3 ~ Kərəsåspa, Parśu 8.6.46 ~ OP Pårsa ‘Persian’, Paktha 8.22.10 (mod. Pashto, Paktho), Varo Susåman 8.60.18 (with unusual Sandhi), Arśasåna 8.12.9, 2.20.6, etc., Anarśani 8.32.2 ~ Iran. əršan-? All such names, if Iranian, belong to pre-Iranian tribes that spoke a dialect close to the one that later developed to E. Iranian.”

Witzel (1995) also sees Iranians in the older books 2 and 6, the conquered enemies of Indra, Drbhika, the tribe of Derbices who, depending on the source, were later either attested somewhere around northeastern Iran or Afghanistan. The hero Turviti (Avestan Tauruuaẽiti) of books 2 and 4 and the Dasa Arsasana of book 2 are interesting contrasts. The Parthava Abhyavartin Cayamana of book 6 who is noted to have a fully Indo-Aryan name is also considered one of the Iranians by Witzel. Most important of those is Prthusravas Kanita whose name is Indo-Aryan and whose patronym is Iranian, specifically found later in Scythia minor where it is the name of a certain king which begs the question: was this Kanita or any of these east Iranians the ancestors of the later Scythians?

There is another controversial but significant term related to the Daevas that appears in both the Gathas and Rigveda, especially in the western hymns, adhrigu ie not dhrigu. Many different translations for drigu exist, usually denoting someone powerless, poor, or weak and an identification for the followers of Zarathushtra. While Lubotsky considers it a loanword from BMAC, it was originally believed to have been borrowed by Iranians from the Indo-Aryans. However, George Thompson (2002) presents some great arguments against that since the word appears in hymns composed in the west and proposes that the base word is the Avestan drigu, who is an enemy of the deva while adhrigu is the deva or their ally. He concludes: “Second, the distribution of the Vedic word, which is largely restricted to the environs of the 8th book of the RV strongly suggests that significant contact between an eastern Iranian language, Avestan, and a NW branch of early Vedic has played a determining role in the semantics of this word-pair.”

It is very clear that some of these Iranians have adopted Indo-Aryan culture or names much like Burrow suggested decades ago but the evidence of Iranian names throughout the old Rigveda does not support separate migration waves since they appear even in the earlier books. Not only that but it seems Zoroastrian influence was expanding when the Rigveda was being composed. The Zoroastrian tradition includes many stories of the Kavi Usana (Usana Kavya) being a stubborn fool and falling under Daeva as well as Keresaspa. It seems from both texts that the Daeva cult was the one associated with the elites in South Central Asia among Iranians, Indo-Aryans and Dasas. All of this leads to more questions: why did the Iranians adopt Indo-Aryan culture and could the Mitanni and Sindoi be among the Iranians who adopted Indo-Aryan culture?

 

The Rigveda: Kulturkugel vs Migration models

If we are to follow the older work of Witzel, the Indo-Aryans showed up suddenly to Afghanistan changing the name of almost every river then somehow forgot everything about their original steppe homeland, then went to India in the next century but did not erase nor forget this Afghan river? They also did not erase the names of other rivers in central and eastern India.

The 200 years for the Rigveda is highly problematic; the only evidence of this five generations of poets is the Anukramanis which is much later than the Rigveda. In fact, there could be gaps between all the books and hymns of the Rigveda which can explain the two homelands. Witzel’s later date for the entrance of the Indo-Aryans changed from 1900 to 1500 BCE after archaeology proved that the Harrapan civilization survived till much later, but now it is 1250-1000BCE which goes against all his “bronze age Rigveda” work, and his previous statements. For example, his 2000 chronology:

1. early Rgvedic period: c. 1700-1450 BCE: RV books 4, 5, 6;
2. middle, main Rgvedic period, c. 1450-1300 BCE: books 3, 7, 8. 1-47, 8.60-66 and 1. 51-191, most probably also 2; prominent: Pūru chieftain Trasadasyu and Bharata chieftain Sudås and their ancestors, and
3. late Rgvedic period, c. 1300-1200 BCE: books 1.1-50, 8.48-59 (the late Vålakhilya hymns), 8.67-103, large sections of 9, and finally, 10.1-854, 10.85-191; emergence of the Kuru tribe, fully developed by the time of Pariksit a descendant of Trasadasyu.

His constant contradictory statements and shifting goalposts are a result of a flawed methodology based on preconceived notions. Under strong rulers, 200 years is enough to modify or erase the collective memory. However, these Aryans were far from unified and were continuously warring against one another; they did not even agree on the pantheon yet both Iranians and Indo-Aryans describe similar homelands and cultures. Witzel (2003) also agreed with Kuiper: “between the arrival of the Aryans … and the formation of the oldest hymns of the RV a much longer period must have elapsed than is normally thought”. He assumed a long acculturation period between the Aryans and the Dasas, but with the genetic evidence (no Steppe ancestry in Aligrama_IA: I8219, 8246) this supposed acculturation period could’ve been pretty late and probably in Iron Age India making the Rigveda contemporary with the Median expansion, but still not knowing of iron tools, and the Atharvaveda with the Achaemenid empire and somehow we have to account that the older Yashts predates the Iranian migration to the west.

Parpola’s multiple wave model, if it has evidence, solves most of these problems. The non-Vedic Dasas came from the steppes in the late 3rd millennium and took over BMAC and then expanded all over SCA Indo-Aryanizing almost everything. They (or the Aryans) invaded Hissar IIIB and by 1760 BCE they reached the Near East and started taking over forming the Mitanni empire. The Dasas change the hydronomy of South Central Asia over time and absorb the non-IE elements and thus when the Aryans arrive they reach an Indo-Iranianized world and those 2-4% foreign words in the Rigveda are filtered through the Dasas who absorbed more of the native language before they are absorbed by the Aryans which explains why more foreign words appear in the later hymns as they expand further away from the Indo-Iranianized Dasa network into non-IE India. After that (or before? It isn’t clear as Parpola has king Sudas defeating Yaz I Saka on the way to India) the Iranians show up. Whatever cannot be explained by the traditional migration theory can be explained by an elite wave or a migration here and there.

However, the reality is that these “pre-Scythians” of the Rigveda are entirely BMAC derived and they moved from Turan to the Steppes and we have evidence of that in the J2a1h2 Taldysay_MLBA2 I4794. There is no textual, archaeological, or genetic evidence for the Andronovo homeland; it is simply untenable. The most parsimonious explanation is that the Proto-Indo-Iranian homeland was in South Central Asia. But before looking at which culture may have spread this language we should see if the Rigveda is dateable.

 

“Chariots” of the Rigveda

“The first appearance of thundering chariots must have stricken the local population with a terror, similar to that experienced by the Aztecs and Incas upon the arrival of the ironclad, horse riding Spaniards” – Michael Witzel

We are always told how the ratha, which is always defined by Witzel as the spoked wheeled horse-drawn chariot, is the most important weapon of the Aryas and what allowed them to take over and impose their language and culture in South Central Asia. Indeed, Parpola believes that the Dasas took over BMAC before the invention of the spoked wheel chariot by the Aryas who later defeated them. Yet, the war booty from the Dasas is 10 chariots and horses in book 6 (47.22-25). Elsewhere (X.102) Mugdalani and Mugdala race in a chariot pulled by bulls. Do these represent the Altyn Depe carts and the BMAC bull racing brought up by Parpola (2015)?

But elsewhere in the Rigveda, there is evidence of chariots pulled by donkeys (III.53.5) translations of Jamison and Brereton(2014) below:

Drive yon, bounteous one, and drive hither. Brother Indra, in both places there is a goal for you, where there is a resting place for your lofty chariot and unhitching for your prizewinning donkey.

So was the ratha like the Sumerian carts? But the biggest surprise of all, the Asvins, the supposed Indo-European divine twins, are associated with the donkey!

Where are the three wheels of your triply turning chariot, where the three seats which are in the same nest [=chariot box]? When is the yoking of the prizewinning donkey, with which, Nāsatyas, you drive up to the sacrifice (I.34.9)

Hitch the donkey to the chariot whose parts are solid, o you who bring bullish goods, to drink of the honeyed soma. (VIII.85.7)

The two fallow bays (of Indra) and the two dappled mares (of the Maruts) have become your yokemates. The prizewinning (horse) has taken his place at the chariot-pole of the (Aśvins’) donkey. (I.162.21)

The association of the Asvins with donkeys and agriculture (I.117.21) especially when the focus of the Rigveda is on the priest-warrior elite made me question many of the accepted conclusions. Parpola (2015) believes that the divine twins do not go back to PIE but spread with the chariots of Sintashta (I’d say the divine twins cannot be Proto-Indo-Iranian deities let alone Proto-Indo-European, their rejection from the Iranian pantheon while others like Mithra were reintroduced is quite telling).

Looking at the Rigveda in 10552 verses there are very few ratha based names, Srutaratha, Priyaratha in book 1, Svanadratha in book 8, and Bhajeratha in book 10. In books 2-7 they only appear in book 5 (Srutaratha, Rathaviti, Saucadratha) and once in book 4 (Citraratha) in what is argued to be a redacted hymn (IV.31). This is also seen in the Younger Avesta where there are only 6 ratha based names. This is contrasted by the Near Eastern Aryans as Parpola notes that the Mitanni mostly have horse and chariot-based names: “To conclude, the culture and proper names of the Mitanni Indo-Aryans were dominated by the horse and the chariot. Indeed, the Mitanni Indo-Aryans seem to have been the prime motors in the introduction of chariotry into the West Asian warfare around 1500 bce.”

The Rigvedic ratha could at times be a solid wheel ox cart at others a Mesopotamian donkey cart or a Hissar IIIB proto-chariot or even a spoked wheel chariot. This is without counting some of the allegorical and mythological rathas without wheels or with seven wheels…etc. Since the line of myth and reality is blurry in the Rigveda the data on ratha is simply inconclusive and should not be used to hard date the Rigveda as is usually done.

 

Camels of the Rigveda

The Bactrian camel was an important animal for the Iranians where several gods are associated with it including the prophet of Zoroastrians. Kuzmina (2007) considered the camel as one of the ethnic markers of Indo-Iranians and one of the reasons for the Andronovo homeland, “The appearance of the Bactrian camel as a draught animal in Baluchistan is probably connected with Andronovo influence. On the post-Harappan settlement of Pirak camel bones and clay camel figurines, including figurines with holes for attachment to vehicles, and horses and mounted horsemen were found. Another important feature of the Andronovo economy was the breeding of the Bactrian camel”. Since most archeologists believe the camel was brought from BMAC to Pirak, Witzel suggested that the Dravidians brought the horse to Pirak since they have a different word for horse than IA but also noted that their word for camel is Iranian.

On closer inspection, it seems that the camel was not an important animal to the early Indo-Aryans. It is mentioned only in 3 hymns once in book 1 (138) and the rest in book 8 (6,7,46). In all but one of these the camels are gifts from the Iranians Kasu lord of the Cedi, Tirindira of Parsu, and Prthusravas Kanita with the last one being the non-IE Dasa Taruksa Balbutha. It is hard not to connect Iranians with the domestication of the camel especially when considering that there is evidence in Mallory (1997) of the IE root of ustra. The lack of camels in the Rigveda has been used by the OIT crowd to date most of the Rigveda before the domestication of the camel (2500-2000BCE). But it should also be noted that camels were rare in the urban centers of South Central Asia: “Moreover, the scarce percentage of bones found in the Iranian and South Turkmenistan settlements should not be construed as evidence against the domestication of the animal, as the camel is used as a form of transport outside the inhabited center, being associated with nomadism” (Bonora, 2020). By this interpretation, the Indo-Aryans/Indra cult could have been in charge of the urban centers while the Iranians could have been semi-nomadic as was the case with Zarathushtra who preached from one land to the other but we will see if that is true or not later.

However, since most of the Rigveda was composed around the Indus the earliest evidence of the camel is around 1700BCE in South Asia. Adding Witzel’s favorite 5 generations we get an approximate date of 1900-1700 BCE for the Rigveda which isn’t far from the original 1900-1200 BCE date. In my opinion, a date around the 4.2ky event (2250-1800 BCE) matches very well with Indra battling the demon of drought being the main theme of the Rigveda. The Rigveda would have then been collected, ordered, and redacted over several centuries with interpolations here and there where the end product would be in a language similar to that of the Atharvaveda.

And just like in archaeology, there is nothing explicit in the Rigveda that should associate the Indo-Aryans with the steppes. It is quite evident that the composers of the Rigveda and Avesta had no idea where they came from originally, which matches well with Erdosy’s (1995) statement on a late third millennium to early second millennium entrance to South Asia: “Indeed, if one accepts that the migrations of Indo-Aryan speakers into South Asia already entered the realm of mythology at the time of the Rgvedic hymns, and that the latter were composed from ca. the 15th century B.C. onwards, the chronology suggested by the archaeological evidence already makes perfect sense”. If the Rigveda was composed around 2000 BCE then the entrance of the Indo-Aryans would have been even earlier.

There is evidence of extensive contact between Iranians and Indo-Aryans from both the Avesta and Rigveda. There is even evidence in the Rigveda that Zoroastrianism was spreading in the west. The Indra cult, while younger than that of the Asuras, seemed to have faced much opposition but ended up being the cult of the elite in South Central Asia among Indo-Aryans, Iranians, and Dasas. The Aryans who appeared in the Near East followed a similar religion to that of the Indra cult with the gods of the Mitanni being entirely the same as that of Indo-Aryans of the Rigveda. Dating this religion is quite problematic, since Indra appears in the Near East at the end of the 3rd millennium it should be dated to 2200 BCE at the latest.

 

Religion and Archaeological Culture of the Indo-Iranians

Many proponents of the Andronovo Indo-Iranian homeland follow reconstructions of Wilhelm Rau whose interpretations in the words of Edwin Bryant “are not explicit in the text”. He is the person many supporters of the steppe theory go to when it comes to the Rigvedic culture, who paints the Indo-Aryans as these migrating nomads in search of booty who never lived in villages and did not know of cities. How convenient, a text composed without a doubt in South Central Asia somehow matches the archaeology of a land far to the north. Even to this day, there are some like Lubotsky who are proponents of this nonsense. Of course, that image is false even if we take Andronovo as the Indo-Iranian homeland, because of BMAC, the culture they traded with.

So what is the story behind these blind and deaf Aryans who have never seen or heard about a city? It turns out that earlier the Aryans were seen as the destroyers of the Harappan cities (pur) however when the Rigveda was searched through it was discovered that the Aryans themselves also lived in purs and so it was decided that the purs of the Rigveda were not cities or forts by any means but rather some mud ramparts for nomads. The simple truth is that if the Rigveda was composed anywhere in South Central Asia anytime in the Bronze Age or the Iron Age based on every archaeological evidence the Aryan must have known of forts and utilized them, “It is important to note that the BMAC-type citadels and manors survived until the Yaz I–related cultures of the Early Iron Age, for instance at Tillya Tepe in northern Afghanistan, and indeed until medieval and modern times” (Parpola 2015). There has to be an explanation for this cognitive dissonance and it can be seen in Kuzmina’s works, where the Aryans are not from a different dimension, and the Avesta and Rigveda were composed in South Central Asia but they are implied to reminisce about the good old days of the steppes thus the perfect and ideal world that is in the Indo-Iranians texts represents Andronovo which somehow ends up making these texts the product of the steppes not of South Central Asia.

Fortunately, the latest translation of the Rigveda by Jamison and Brereton does not have much of this nonsense even though they do accept that the Indo-Aryans were semi-nomadic and it is evident in their translations (eg. grama is a nomadic band instead of village). In my opinion, they describe the hymns of the Rigveda pretty accurately: “First, the R̥gveda represents the continuation of an elite tradition also attested in the Avesta and therefore quite ancient. As such, it reflects the religious practice only of the upper strata of Ārya society. Second, it is primarily a collection of liturgical hymns for use in the soma sacrifice, surely the most prestigious ritual of the period but still only one kind of ritual, representing a particular and limited set of religious concerns. Finally, the soma sacrifices were sponsored and performed by socially elite men, and they reflected the religious concerns of these men. The text did not directly address the religious lives of women or of other social classes nor indeed even other aspects of the religious lives of elite males. Thus, while the R̥gveda is a sizable text and from it we can derive a great deal of information about the soma rite and about those who participated in it, we are still dealing only with a segment of Ārya religion and society”.

 

Settled life

Despite it being composed for the elite, and while never the focus there is clear evidence of agriculture from the oldest to the latest books in the Rigveda, where evidence of grain (I.16, III.6, IV.7, V.21, VI.30..etc ), granaries (II.15), barley (I.21, II.19, V.85, VII.3,..etc), and plow is abundant which gives an image of an agropastoralist society. Weirdly enough there is one peculiar hymn in book 4 (57) that is very different from the rest of the Rigveda dedicated to agricultural deities that somehow made its way into the Rigveda with its importance noted by Jamison and Brereton: “Hymns like this give us precious glimpses into everyday life and the technical terminology of particular profession”.

Other South Asian indicators are missing from the Avesta, such as water buffalos that appear in every book and rice porridge in book 8. Boats have been downplayed by the steppe crowd ever since they have been used to connect the Rigveda to the IVC. Boats and ships are mentioned in every book except the fourth one and not just for crossing rivers as some claim but also for the sea (I.25.7, VI.59.3, VII.88.3). Most importantly the Aryans were not just the destroyers of forts but they lived in them too (X.102.8):

Make a pen (for it), for it gives drink to your men. Stitch (for it) (suits of) armor, ample and broad. Make (for it) fortresses, metal and unassailable. Let the beaker not leak: make it solid.

Even the Deva gods are associated with these forts not just their Dasa enemies, in I.166.8 the Maruts have a stronghold with a hundred coils. However, it is Agni the houselord who is mostly associated with these forts:

Who give gifts and rewards of horses with a desire for great fame— rescue them from difficult straits with your rescuers and with your hundred fortresses, o youngest one

Fire worship is a core part of IE religion where there is specifically a concept of eternal flame not only in Indo-Iranian religions but also in Europe as seen with Vesta and Hestia. Therefore it should not be a surprise that the god of fire is associated with settlements or how else would that eternal fire be tended, on Yamnaya wagons?

Tying up the Rigveda and Avesta with Andronovo will probably never stop since Kuzmina somehow saw evidence of Andronovo houses in the Avesta:

“In the Avesta, in the Ardvīsūr Yašt (5.101), dedicated to the goddess Anāhitā, it is said: “At every stream there is a solid built house, it is light, with a hundred of light openings, well-made, with a thousand pillars, firm, with ten thousands of supporting pillars.” From all the various house types that are archaeologically attested in Eurasia during the 2nd millennium BC this description of Anāhitā’s house is closest to the large Timber-grave/Andronovo house.”

To support her theory she brought up one of the most controversial Yashts, which is believed to have been changed during the Achaemenid period, where the non-Iranian goddess, Anahitis, merged with the river and took over the role of Apam Napat the Iranian counterpart to Varuna and perhaps other aspects of Avestan deities. In my opinion, whether these syncretic Mesopotamian aspects of Anahita were being incorporated during the younger Avestan era or separately later during the time of the Achaemenids it is very likely that this goddess was a favorite of the Persians, especially of Artaxerxes II where he breaks tradition and invokes her before Mithra. But where is this Andronovo house with a thousand pillars and windows? I have a feeling that Kuzmina isn’t showing the whole picture so let’s take a look at the following line.

101. ‘Who has a thousand cells and a thousand channels: the extent of each of those cells, of each of those channels, is as much as a man can ride in forty days, riding on a good horse. In each channel there stands a palace, well-founded, shining with a hundred windows, with a thousand columns, well-built, with ten thousand balconies, and mighty.
102. ‘In each of those palaces there lies a well-laid, well-scented bed, covered with pillows, and Ardvi Sura Anahita, O Zarathushtra! runs down there from a thousand times the height of a man, and she is possessed of as much Glory as the whole of the waters that run along the earth, and she runs powerfully.

A monumental pillared building with couches and scented pillows, let’s not beat around the bush this sounds more like a palace, especially the Persian Apadana! This Yasht does not support Kuzmina’s statements in any way, it does the opposite. However, since some elements of this hymn were originally supposed to be dedicated to Apam Napat or *Vouruna, it could also mean that parts of this hymn may go back to Indo-Iranian times.

The two kings without deceit—in the highest, steadfast seat with its thousand pillars they sit (II.41.5)
Where have these companionships of ours come to be, when previously we would have accompanied one another without wolfish hostility? O Varuṇa of independent will, I went into your lofty mansion, your house with its thousand doors. (VII.88.5)

The description is hyperbolic but it seems to be consistent so the building that Varuna’s palace is based on does predate the Achaemenid apadana. Massimo Vidale (2017) in a great paper on the Vara of Yima analyzed the imagery of the Jiroft “handbags” that could be considered a visual reference to the monumental buildings of Mundigak and Tureng Tepe that are embellished with semi-pillars and windows. It turns out these handbags and buildings are a reflection of the Vara of Yima and he boldly concludes: “south-eastern Iran of the second half of the 3rd millennium should be fully included in the sphere of cultural interaction we conventionally define with the term of Proto-Indo-Iranian.”

While the myth of the Vara appears in the late book Vendidad it still derives from a common Indo-European origin, where a king, usually of the dead, rules a hall or castle with abundant food. Initially, I thought these monumental buildings fit better with the palace of Varuna but it seems they might be the same thing after all according to the Rigveda (X.136):

Here is the seat of Yama, which is called the palace of the gods. Here is his pipe blown; here is he adorned with hymns.

Recently, it has been argued that these monumental buildings, of which Tureng Tepe is the largest, have their roots in 2800BCE Mundigak which is around the Harahvati. These buildings spread throughout South Central Asia in the third millennium, in Altyn Depe, Nad- I Ali, Gonur, Hissar..etc. Those in BMAC have evidence of the Soma cult which may have its origin in Margiana: “Fire is also said to have played a major role in these buildings. Many have raised doubts on these statements, but the presence of seeds of hemp and pollen of ephedra in jars cannot be denied and testifies that these installations were dedicated to the confection of hallucinogenic drinks” (Lyonnet and Dubova, 2020). In the burnt building of Hissar IIIB there were weapons, golden beads and diadems, and semi-precious stones around the fire altar but I am not sure if there is evidence of the soma cult.

All of that leads us to the obvious conclusion that the Rigveda and Avesta have nothing to do with the steppes in any way. They are a product of South Central Asia and if anything they do not reminisce about the “good old days of the steppe” but rather of monumental buildings with fire altars.

 

Dragonslayers

Ignoring the theories about the origins of Islam, nowadays most scholars believe that the Islamification of Iran and South Central Asia was a bit later than what the traditional Islamic narrative claims. After this Islamification process much changed in those Zoroastrian areas in terms of art and culture. Coins were struck in the name of Allah, bronze statues and rock reliefs were replaced with Arabic calligraphy, and the Iwan was no longer just the gateway of the King’s palace but rather to the new house of god the mosque that spread all the way to India. A few centuries earlier, the Kushans and Huns spread statues of Buddha across the Iranian plateau. Even earlier, the Greeks formed kingdoms in South Central Asia where statues of the Greek heroes can be seen today from Media to Bactria. Even the coins of the Parthian king of kings were inscribed in the language of the Greeks.

Ancient art can be interpreted in many different ways but in general, everything is looked at with Mesopotamian or Steppe goggles. Every stick and stone found in a kurgan burial is the vajra of Indra and every statue all the way in Bactria somehow depicts Gilgamesh or Humbaba. If there is evidence of Indo-Iranians in South Central Asia then it should be visible in the art.

Bruce Lincoln (1976) reconstructs a myth where the hero associated with thunder, *Trito, kills a dragon in a mountain or a cave and rescues his cattle. While the dragonslaying myth is accepted as a core part of Indo-European religion, Lincoln interprets it specifically as an imperialistic cattle stealing myth. It is hard to connect the Illyunka myth to a cattle raid. In my opinion, there are three variations of the dragonslaying myth, one for each supposed function.

The Mitanni association with the Iranian Grey Ware to BMAC is generally accepted because of this LBA Hasanlu bowl. The bowl depicts a battle of a hero battling a multi-headed monster, with the god above him on a chariot that is pulled by bulls. The hero who is wearing the East Iranian kilt is most likely Trita or his son. An interesting observation is that both the god and the hero are wearing the same type of headwear. I agree with the Mitanni interpretation and believe that this myth aligns more closely with the Indo-Aryan version rather than the Zoroastrian one. In the Iranian version, the main dragonslaying myth is a warrior/kingship one and not associated with the release of the water as in this bowl and in the Rigveda. Another thing that should be considered is that in the Younger Avesta Azi Dahaka plays a different role than the other dragons and that he is more man than beast. He is more like an evil hero much like the Turanian Afrasiab as he worships and sacrifices to the gods.

Francfort (2008) interprets the symbols of this bowl as a mixture of Hurrian and Indo-Aryan with East Iranian (Jiroft to Oxus) influence. He reconstructs the East Iranian myth as follows:

1) the water flows from heaven, provided by bovid (zebus)/goddess
2) the water is kept by the mountain dragon (on Khafaje the bovid is represented as dead and attacked by felines and birds of prey);
3) the hero (eagle or mountain goat that is an anthropomorphized ally of the main goddess in the Oxus Civilization) fights the dragon (snakes and felines on Khafaje) and releases the water that fertilizes

Shockingly this doesn’t sound like an Indo-European myth to Francfort but rather one adopted by the incoming Indo-Iranians. The bird, goat, and snake motif is thus considered non-IE. There are many different variations of this myth with different protagonists so I won’t be going over them all and since the BMAC goddess is missing in southeastern Iran I consider it a BMAC tradition. First, I would like to dissect the common myth and then look at the BMAC goddess. This is not to say that the entirety of East Iran and Turan spoke one language or were of the same ethnicity since the city of Babylon alone had Semitic people, Elamite foreigners, and Hurrians all living in it; it is simply the most likely myth associated with the elites of that region.

The eagle or a hero associated with it fighting the dragon should be no surprise, it is one of the forms of Indra and Verethragna and the symbol of royal glory but here since it can always be claimed that it was borrowed from the natives I won’t argue for an Indo-Iranian connection but rather an older Indo-European connection so that the argument that the BMAC religion is similar to Indo-Iranian is because it was absorbed is nullified. The eagle is one of the forms of the Indo-European thunder god and dragonslayer and his messenger and a symbol of many Indo-European empires, from Achaemenids and Greeks to Hittites whose double-headed eagle comes from Iran. The SPQR standard of Rome depicts an eagle holding Jupiter’s thunderbolt as do many Greek coins. The eagle-snake motif itself is found in Germanic mythology where at the bottom of the world tree the dragon Nidhoggr is chewing the roots, at the top there is an unnamed eagle whose function is unknown. The same can be seen in Slavic mythology where the underworld god Veles takes the form of a snake at the root of the world tree and his nemesis the thunder god Perun is an eagle at the top.

While the goat god may have lost his prominence when many of the IE myths were recorded (see Ir. Mithra who probably absorbed Pusan) the goat or goatman still has many parallels in Indo-European myths, especially in Zeus’ foster brother Pan who helped Zeus defeat the titans but also in Faunus, Pusan and many other Indo-European spirits. The goat is also associated with the thunder god where two goats pull Thor, the lord of goats, and Perkunas’ carts. Perkunas also rides a goat or assumes a goat form in a riddle (West 2007). In Greece, the role of Fereydun’s cow is taken over by the foster mother of Zeus goat Amalthea whose skin is later worn as an aegis. The winged goats also guard the world tree in Scythian art much like in Iran.

The dragons themselves are even more interesting. At times depicted with three heads as in the Indo-European myths and among those, there is a special one in BMAC an ugly demon dubbed by Sarianidi “the proto-Azi Dahaka”. The Azi Dahaka of BMAC is different from the one in Zoroastrianism who is more manlike though Keresaspa does defeat a dragon associated with water. However, there is a cylinder seal of unknown origin now believed to be from Eastern Iran dated to the end of the third millennium that depicts a different kind of Azi Dahaka who is seated as a king in a court playing the role of the snake sacrificing the chopped up zebu cow and that in my opinion is the proto-Azi Dahaka of Zoroastrianism.

Such early depiction of the core IE myth much earlier than the rest of the world (2700BCE) makes it very certain that there was a group of Indo-Europeans present in eastern Iran especially when considering most of the dragonslaying myths of the Near East are almost a millennia later than this one when Indo-Europeans are attested around Old Babylon and the New Kingdom of Egypt (Ra and Apep).

 

The Goddesses of BMAC

A seal depicting the Lion Goddess of BMAC and a golden Scythian bowl from Kelermes barrow 4, Kuban (7th-6th century BCE). Source: Sarianidi, 1998 and Piotrovsky, 1986.

The evidence of goddesses has long been used to argue that BMAC was not Indo-European but at the same time, it is claimed that the religion of the Scythians is very close to Proto-Indo-European despite them mostly worshipping female deities. Nani or Nanaya as she would later be known, the lion goddess of BMAC is known to have her origins from the Near East (Parpola 2015). Being originally non-IE should not be a problem as Indo-Europeans are known to have adopted non-IE gods as well. There are many depictions of her later across Eastern Iran from Bactria to 7th century Chorasmia and I’d argue even in Scythia.

While there are a few Scythian goddesses that are associated with snakes the goddess does not appear in a Western Iranian or Zoroastrian context. Several western Indo-European goddesses have similar portrayals of the subdued snake, Hygeia, Verbeia, Sirona..etc but they probably do not go back to Proto-Indo-European times. However, Danu the mother of Indra’s nemesis, Vrtra, does go back to Proto-Indo-European so there may be a connection between a river/fertility goddess and the dragon in IE mythology. Azi Dahaka of the Avesta only prays to two gods the Yazata-Daeva Vayu and the river goddess Anahita. In Greece, Hera sends two snakes to kill the young Heracles and later the Lernaean Hydra but also in some stories she’s the one who sent Typhon to kill Zeus, not Gaia.

While a Greco-Aryan connection could easily be explained through historical contacts, contacts between Ireland and Indo-Aryans anytime is much harder to explain. And so a healer-dragonslayer and Vrtra-Danu myth has been reconstructed for PIE by John Shaw (2006). He noted parallels in the Irish myth of the Trita-like healer Dian Mecht who killed the dragonlike son of the mother goddess Dannan who along with Ana and Morrigu go back to a *Danu prototype. Here Azi Dahaka’s sinful mother Otak is also considered to be Danu. It should also be noted that the Scythian mother goddess Api is also associated with water.

 

The Identity of BMAC Elites

So who were the people of BMAC? Eagles being related to the gods and snakes and felines as antagonists is common in Iran. Sarianidi proposed that they were Proto-Zoroastrians long ago but the goddesses of BMAC are vilified in the Avesta and do not appear in later Zoroastrian contexts nor in western Iran, the art of BMAC is more related to that of the East Iranians and Scythians. However, if these were Indo-Iranians then there would be numerous depictions of horses, right? Well, just like in BMAC it seems that horses do not play a large role in Indo-Iranian cosmology as seen in Scythian art (see here):

“Ungulates, as a rule, are easier to identify; the most frequent images of them were those of deer, goats, and rams, while other species, including horses and elk—were only rarely depicted. These three groups of animals depicted in Scythian art were probably associated with the three Cosmic horizons; birds with the upper level (that of heavens), ungulates with the middle level (that of the Earth), and beasts of prey, as well as also snakes and fish, which were depicted less often, with the lower level below the earth.”

Iron age Assyrian birdman and BMAC two-headed birdman fighting a dragon and a boar.

Even so, one of the major heroes of BMAC, the birdman, does not appear in an Iranian context and is most likely a Bactrian innovation based on the earlier East Iranian eagle. The birdman appears exclusively with Indo-Aryans both those in the Near East and those who composed the Rigveda (I.164.46). Sarianidi when excavating BMAC following late Soviet chronology considered it to be derived from Syro-Anatolia due to the similarities with the Mitanni. One of the figures that shows up in both BMAC and the Mitanni is the birdman that continues to appear even later in the Neo-Hittite and Assyrian empires. This is generally accepted today: “We cannot exclude interrelations between the BMAC and Anatolia at the beginning of the second millennium, since the Karum Kanesh seals and old Syro-Hittite seals share some common motifs with BMAC, like the doubleheaded eagle with guilloche, the bird-man, and others” (Winkelmann, 2020). So how come this non-IE BMAC or language X or whatever made up stuff is not attested in the Near East but rather an Indo-Aryan language? Parpola (1995) had the right answer: “At this time Assyria was trading with Cappadocia and importing tin from the east. The source of this tin may have been in central and northern Afghanistan (Kandahar and Badakshan), whence the Harappans and the Bactrians appear also to have obtained their supplies… On the other hand, from the 18th century B.C. onwards, north Syrian seals show such a typically Central Asian motif as the two-humped Bactrian camel, which is depicted in the BMAC seals several times…These cultural contacts between the Syro-Hittite world and the BMAC do not prove that the hypothetical Aryan authors of the BMAC came from the west, as suggested by Sarianidi (1993b, 1994), but rather foreshadow the takeover of power in Syria by the Mitanni Aryans and support their Central Asian origin.”

There seems to be enough evidence to support the claim that the Mitanni spread along the tin road. We can even speculate that the reason for the dominance of the deva cult was a side effect of this Mundigak tin road, and both BMAC and the IVC were getting their tin from these regions.

 

Indus valley

With its hearths and post-cremation urns Cemetery H was considered to be Vedic before it was considered impossible to find a link with the steppes. In contrast to BMAC, unfortunately, every paper out there on the Indus Valley Civilization is contradicted by another one. From the available evidence, reconstructing the rituals and beliefs of one of the largest civilizations is very difficult, to say the least, especially when considering there are differences between Margiana and Bactria in BMAC. In Kalibangan there were seven hearths found in a non-residential area interpreted by the Allchins as ritual hearths. Parpola (2015) considered it Vedic: “In the Vedic ritual, the word dhiṣṇ(i)ya in the strict sense denotes the fireplaces of seven priests officiating in a soma sacrifice; six of them are built in a row in the sitting hall, one is outside the hall on the border of the sacrificial area. A very similar row of seven Indus Valley fireplaces has been excavated on a ceremonial brick platform in the acropolis of Kalibangan”. Yet another interpretation is that these hearths were for cooking and baking. The core issue is that everything revolves around either proving the IVC is the Vedic civilization or it is non-IE, unlike BMAC whereas recently the Indo-Iranian debate has been taken out of the question. It is hard to objectively look at the IVC especially now that the chronology of the surrounding region is now being revised.

There isn’t even anything concrete on what was the model of the Indus societies, were the elites merchants or priests? There have even been some outlandish models for the IVC that have no basis in reality where its people were peaceful egalitarians. So what stopped their warlike expansionist neighbors in Iran and Turan from simply running them over and taking their resources? Most likely because they were a part of the Indo-Aryan network: “The simple socketed axe-adze from Mohenjodaro closely resembles ones from Hissar (Schmidt 1937:P1.52) and Shahi Tump (Stein 1931). At Hissar this axe was found in IIIC context (Schmidt 1937:205). A macehead from Mohenjodaro, dated to ca. 2000 B.C. (Piggott 1947:31), also is typologically similar to one from Hissar IIIC (Schmidt 1937:P1.52). Similar maceheads are known in Luristan, and dated there to ca. 1400 B.C” (Lamberg-Karlovsky, 1967). Parpola also argues that the Indo-Aryans have adopted this axe from Hissar and its appearance in a Kassite context, where Indo-Aryans names are attested, does add evidence to that.

Chah-i Torogh 2 and Mohenjo-Daro Priest Kings. Source: Vidale, 2018.

Weapons are not the only connection between IVC and Iran but rather perhaps the elites of both regions are connected. The Priest King of Mohenjo Daro is now believed to have been a symbol of BMAC and Sistan rather than the IVC with Kenoyer arguing that the Priest King belongs to IVC minorities or foreigners who brought it with them from Kandahar. Interestingly it may also have been found in an elite context: “Finally, it is important to note that at least part of the statuettes’ fragments at Mohenjo-Daro were found in “monumental” contexts and near elaborated accesses, although in secondary contexts of deposition, an evidence that supports their peculiar, if temporary, ideological relevance” (Vidale, 2018). Kenoyer’s (2004) timeframe in my opinion is very solid: “Therefore, it is not improbable that some communities referred to in the Vedas were passing through or living in the regions controlled by Harappa during both the Harappan (Period 3C 2250–1900 BC) and the Late Harappan times (1900–1700 or 1300 BC)”. Indeed, the Deva cult spread across the Tin Road and perhaps originally formed in Mundigak with later contributions from both BMAC and the IVC. In the regions around the Indus, it would be associated with the elites around 2200BCE with Rigveda being composed a few centuries later around 1900BCE in the post-urban phase, following perhaps a Sanskritization process. While I do agree with Witzel’s claims that there was some sort of a Sanskritization process, I do not think it was necessarily established by the Kuru Kings following the Battle of the Ten Kings since the battle is very vaguely described and perhaps obfuscated. Still despite this Sanskritization supposed process many of the Indo-Aryan languages spoken in India today may not directly descend from Rigvedic Sanskrit.

It doesn’t matter what the masses of the IVC believed in since we have evidence of the same elites who had Soma fire rituals and believed in Dragonslayers and three-headed dragons and that is enough for an IE presence. After all, this is the logic the steppe crowd follows: “It is also clear that the Rigvedic Ārya employed some sections of the local populations, i.e. the lower class, called Śūdra since RV 10.90, for agriculture (ploughman kīnåśa, RV, see Kuiper 1991, Witzel 1999a,b), and probably for washing (AV+, Witzel 1986), and especially for pottery” (Witzel, 2001). Even so, not every Indo-Aryan group was Vedic and most likely not everyone was IE speaking around the Indus.

 

The Mitanni and the First Attested Iranian

Bactria during BMAC can be seen as non-Zoroastrian Iranian with perhaps elites of the Daeva cult who deified the Garuda birdman. Evidence of Indo-Aryan toponyms may be seen with Vaksu the later river deity of Bactria. The currently accepted theory on the Mitanni is that they came with Hurrians from Northwestern Iran around Urmia, where the Hasanlu bowl is found, in the 18th century BCE with their ultimate origins in Tepe Hissar III. A question that is sometimes brought up is could these supposed Proto-Aryan features be due to an Iranian influence?

Looking at some of the Indo-Iranian names in the Near East most of them are clearly Indo-Aryan, especially the deities. However, some of them could be seen as Iranian but are reconstructed as Indo-Aryan, some examples (Gentile, 2019): Ar-ta-am-na, O.Pers Artamna (Artamnes), IA *rta-mna and i̯a˗aš˗da˗ta , Yašdata, O.Pers Yazdata (ia˗iš˗da˗da), IA *yaja-data. However, Kosyan (2006) brings to light the earliest attestation of an undoubtedly Iranian name in western Armenia in the 14th century BCE: Uit(t)arn[a(-)…], “Nevertheless, irrespective of what the missing characters are, this name is definitely the Iranian Vidarna-, a masculine Persian proper name meaning”. It is hard to know with the available data whether they were among the Mitanni elites or not but this pushes back the first contact of Iranians with the Near East.

The story of the western Aryans does not end with the Mitanni but rather they survive into the iron age. In Eastern Europe, some loanwords are related to Indo-Aryan rather than Iranian which made some linguists opt for a Proto-Indo-Iranian and Proto-Balto-Slavic clade. Herodotus (V.9) claims that there were colonists from Media who settled beyond the Danube called the Sigynnae. There is archaeological evidence for this movement, “According to a widely accepted, but unproven, theory, the Siginni are thought to be an Iranian people affected by the Scythian invasions. However, we are not able to determine their language or ethnic affiliation. They may be considered as an Iranian or pre-Iranian tribe. The Siginnian presence, whatever may have been the ethnic affiliations of this people, brought about the ethnic and cultural transformation of the Hungarian plain” (Olbrycht, 2000). The Sigynnae relationship with the Sindoi and the idea that some of them migrated together (Roger, 2007) makes it less likely that they were Western Iranians and more likely related to the Mitanni and Indo-Aryans of Northern Iran. Of the Sindoi, Hesychius claims: “Sindoi ethnos Indikon”, this can be confirmed by the hydronyms of the Kuban (Kubha) and Sindes rivers. Even the neighbors of the Sindoi the Maeotians may have had Indo-Aryan speaking elites as Sergey Kullanda (2014) relates their ethnonym to Mitanni.

We have a recent sample in the northern fringes of Scythia, UKR036, archaeologically assigned as a Scythian though more likely an acculturated local chieftain considering the geographical location and the typical native Eastern European Y-DNA R1a-Z645 and significant Transcaucasus-related ancestry.

Target: Ukraine_Kyiv_Scythian_RightDnipro_IllThr_2_EIA:UKR036
Distance: 2.1083% / 0.02108268 | R3P
38.8 Armenia_IA
31.4 Mongolia_EIA_2
29.8 Ukraine_Kyiv_Scythian_RightDnipro_IllThr_EIA

Target: Ukraine_Kyiv_Scythian_RightDnipro_IllThr_2_EIA:UKR036
Distance: 1.9076% / 0.01907649 | R4P
38.0 Armenia_IA
29.4 Mongolia_EIA_2
16.8 Ukraine_Kyiv_Scythian_RightDnipro_IllThr_EIA
15.8 Russia_Srubnaya

In light of these connections, Proto-Indo-Aryan on the steppes is not necessary as these supposed Indo-Aryan words could easily be explained through contact with the Western Aryans. An alternative scenario would be that these Sindoi migrated, with a BMAC related group.

 

The Proto-Indo-Iranian Homeland

Accepting the Indo-Iranian presence in BMAC leads to two scenarios:

1. Proto-Indo-Iranian comes from BMAC.
   In this scenario, Proto-Indo-Iranian would be the lingua franca of the BMAC-IVC network. This type of scenario has archaeological support (Ratnagar, 2020): “One may argue that the exotica found in Mohenjo-daro and at Altyn or Gonur were a matter of random gifts or incidental exchange, yet valued objects could have been given as gifts at marriages, or as diplomatic gifts between elite families for reasons of prestige or political alliance, or even as tribute or homage”. Indo-Iranian would differentiate into Proto-Indo-Aryan and Proto-Iranian around the collapse of the mature phase of BMAC (circa 1800-1700 BCE). However, the attestation of Indara in the Gudea inscription and the appearance of the Deva worshipping maryannu already in the 18th century makes it unlikely.
2. Proto-Indo-Iranian is an early split.
   The more likely scenario Proto-Indo-Iranian splits after Anatolian or Tocharian and differentiates before the formation of BMAC. The IE Dasas who are the ancestors of the Dahae and Khotanese come from this region. The non-IE Dasas (eg. Taruksa Balbutha and Ilibisa) either were a minor group, the previous natives, the people of the region beyond the Oxus, or neighbors of this system.

There are constraints for both scenarios. It is difficult to determine how exactly the Indo-Aryan reached the Indus knowing the material in the Vedas. Considering that the Mitanni carried a BMAC-related package to the west and that BMAC-IVC hybrids sites are seen in Sibri/Mehrgarh/Nausharo/Quetta/Dauda-Damb it is very tempting to say that Indo-Aryan languages were imposed by BMAC elites in the late Harappan period. However, it can still be argued that the BMAC elements were brought by Iranians who adopted the Vedic culture.

 

Reverse Engineering BMAC and Proto-Indo-Iranian Migrations

Much like the later Yaz I culture BMAC originates in Kopet Dagh Piedmont to the east where before its new center would shift to Gonur. BMAC would expand in a “khanate” model, brought up by Lamberg-Karlovsky, where a tribal leader, just like in most of the history of South Central Asia, would be the political head of the forts. The aggressiveness of BMAC and its violent expansion to the west has long been considered with the destruction of Hissar IIIB being attributed to it. This view has recent support with the evidence of a goblet depicting a defeated southeastern Iranian (Vidale 2017). However, there does not seem to be strong evidence of violent expansion in the Indus Valley, perhaps a sign of an alliance.

These forts were surrounded by pastoralist groups whose ethnicity has been discussed “It is not yet clear, however, whether the mobile groups were related to steppe groups or to local BMAC herders in a dimorphic-like society” (Lyonnet & Dubova, 2020). The appearance of Andronovo pottery has long been used as evidence of migration of the Indo-Iranians but aDNA has answered the question about the origins of the nomads as there is continuity of BMAC well into Yaz. The pastoralists who carried chlorite artifacts outside of the forts must have been some sort of underlings to the “khans”.

 

Proto-Indo-Iranian Migrations

To trace the movement of Proto-Indo-Iranians the Namazga culture has to be looked at, since BMAC itself derives from the earlier Namazga culture with an extra Mundigak influence (Lyonnet and Dubova, 2020). Either PIE was native to Namazga and developed into Proto-Indo-Iranian or was brought up by migrants to Namazga. The 4th millennium Uruk and “Proto-Elamite” expansions can be excluded as there is no direct evidence of their influence in northeastern Iran and Kopet Dagh, but trade cannot be excluded.

The most impactful migration is related to that of the Iranian Grey Ware people, who a long time ago were believed to be invaders from the steppes 3000 BCE. The origin of this grey ware was in Tepe Pardis, just to the south of the Alborz mountains in the 6th millennium BCE. The grey ware spread gradually in the 4th millennium among the elites of northern and northeastern Iran and western Turkmenistan (Bonora & Vidale, 2013). However, it is after 3400 BCE in Hissar that the evidence of violent conflicts and destruction occurs. In the late 4th millennium a complex society was formed in Hissar, “Locally, the craftsmen, farmers and herders, and the elite administrators would be members of a socially hierarchical but economically interconnected community whose basic sustenance depended upon a prosperous agricultural economy” ( Gursan-Salzmann, 2016), and much like the later Gonur, the elites were buried in cist graves.

During this period some sites related to the Anau culture were abandoned and Ak Depe was even burnt. This all relates to evidence of a migration from Northeastern Iran to Ashgabat (Kircho, 2020). While grey ware is present in Turkmenistan (3200-2800 BCE) it is still not the dominant pottery in every site, unlike in Namazga IV where the period is characterized by a grey ware pottery variant. It was in this period that a migration from Turkmenistan related to Geoksyur/Namazga III to Shahr-i Sohkta, Mundigak and Mehrgarh. Some of these migrations might have been accompanied by violence, as there seems to be evidence of burning in Mundigak IV and SiS. However, it is unknown if the destruction of Kot Diji is related to the migration of these groups. In this case the formation of the IVC would also be associated with this network.

The most likely prestige language and lingua franca of this network is Proto-Indo-Iranian. Gian Luca Bonora’s thesis on the linguistic reconstruction of prehistoric Balochistan examining terms related to botany, agriculture, livestock, tools..etc, he concludes most of them have clear Indo-Iranian origin.

After 2800 BCE the unity of this network slowly broke with the appearance of the first “proto-towns” signaling regionalization (Wilkinson, 2014). The differentiation of Proto-Indo-Iranian in this case should be no later than 2500 BCE. Archaeologically, the areas where the Indo-Aryan influence is attested have evidence of contacts with BMAC, which may have begun 2400-2200 BCE. This is not surprising as the soma cult, which is associated with the Devas, first appeared in Gonur. In this case, if BMAC was not Proto-Indo-Aryan, likely, the origin of some of the rituals and beliefs of the Deva based religion of the Rigveda should be sought there. In this case, Mundigak (I predict it to be very similar to the IVCp samples) would be where Proto-Indo-Aryan split and entered India.

 

IRANIAN MIGRATION

The Iranian migrations have been problematic to explain in both terms of textual and archaeological evidence. Unfortunately, the Avesta is less understood than the Vedas where there is an established relative chronology in terms of the texts as well as the evolution of some social occupations (eg. marya > rajanya >kshatriya). On the other hand, there are some very unlikely theories on the Avesta, such as the one that claims the Younger Avesta predates the Gathas that were supposedly composed in the 7th century BCE. Following the more likely Older Avestan > Younger Avestan approach Jean Kellens (1987) states: “The Old Avestan part is even more ethereal. Not only are there no allusions to historical events or geographical places, not even mythical ones, but for the language as for the religious concepts, there is a complete break between this part and the young Avesta, more recent by several centuries, fore which nothing at all is known“. He puts the era of Zarathustra to 1200-800 BCE by comparing the evolution of Old Persian. However, in recent works, it has been shown that the written Old Persian of the Behistun inscription is much older (and may have never been spoken) than the Persian language spoken in the Neo-Elamite and Achaemenid periods (Henkelman, 2011) as the spoken language already had Middle Iranian features (Tavernier, 2011).

What we know from the Gathas is the following: “His hymns suggest that Zoroaster belonged to a settled community, and perhaps his tribe had already occupied a part of Khwarezmia before he was born. If so, it may have been tradition (and probably the conventions of religious poetry were highly conservative) which kept his imagery still strictly pastoral” (Boyce, 1996). Of course, pastoral imagery was very common in Mesopotamia, Egypt, and many other settled civilizations. The Gathas describe attacks on peaceful settlements (Y.48.10), with the attackers being raiders following usij, karapan, and kavi priests. Boyce’s claim of the Daevas being related to the Scythian deities is unlikely as the Scythian religion developed much later around the Dnieper. There are no mentions of Enarees nor are the Scythian deities known in the Younger Avesta. According to Jamison and Brereton: “the uśíj priests were masters of ritual action, while the kaví was the master of words and knowledge”. In this case, the enemies of Zarathustra are the Indo-Aryan warriors, probably the mairyas of the Younger Avesta, who went as far as Anatolia and Indo-Aryanized Iranians of the Rigveda putting the Zarathustra in a similar milieu as the Rigveda.

What is more important is the evolution of the warriors in the Avesta: “In Zoroaster’s hymns these were called by the following terms: nar, literally “man”, that is, the fighting man or warrior…The warrior is usually known as rathaestar “chariot rider”, a term evolved evidently after the Iranians had adopted the war-chariot instead of fighting on foot…but the vocabulary of his poetry still reflects an older state of affairs, when probably only the weak and old travelled in heavy ox-drawn carts, and men would have made their way, and fought, on foot” (Boyce, 1996). Knowing without a doubt that chariot warfare began in the 18th century BCE at the latest in the Near East and Zarathustra’s homeland was closer to the area where the spoke-wheeled chariot was invented, this puts him at a very early date.

Currently, there is a gap in the data of the Post-BMAC/Pre-Yaz or Final Bronze Age (~1800-1500 BCE). The reason for the changes in socioeconomic and social structures is unknown and many sites are abandoned (Luneau, 2020). The previous religious elite disappeared and the rise of Proto-Zoroastrianism has been attributed to this phase in BMAC by Sarianidi. Evidence of a multicultural society and trade with Andronovo-related groups is well attested. However, as Parkhai_LBA (excluding the outlier), Dzharkutan1_BA, Bustan_BA, and Sumbar_LBA show no steppe ancestry, this multiculturalism is not a product of Andronovo-related groups. Then the more likely cause of this multiculturalism is the spread of some type of Zoroastrianism and perhaps the Indo-Iranians who moved to the steppes (Taldysay and Kyzlbulak) rejected this movement.

The Early Iron Age (post 1500 BCE) sees the formation of Yaz I and the return and expansion of highly organized Grey Ware societies of northeast Iran as the sites may have been abandoned after 1800 BCE or replaced by BMAC settlements.

“The disintegration of these settlements resulted in the diffusion of the Late Bronze Age cultures and their subsequent transformation into what is known as the Early Iron Age in adjacent areas. The desertion of important Late Bronze Age sites of the north-east may have caused the spread of “grey ware cultures” all over the area, from south-western Turkmenistan to the central regions of the Iranian Plateau. The origin of such a phenomenon is still a mystery, but an environmental crisis combined with disequilibrium in the network of trade in north-eastern Iran could well account for the eventual decline of the Late Bronze Age settlements” (Mousavi, 2008)

In this case, the common scenario of the Yaz I being related to the Iranians will be accepted. Cuyler Young saw the spread of Early Western Grey Ware (1300 – 1000 BCE), which falls within the accepted dates for the entrance of western Iranians, from Northeastern Iran to Hasanlu as related to the migration of the Western Iranians, and the common objection that this grey ware was not brought by steppe nomads is irrelevant for this model since we expect continuity from the copper age.

The Median forts of Nush-i Jan, Godin Tepe, and Tepe Ozbaki show similarity to the the citadel of Ulug depe which goes back to the BMAC forts. Adding to that in Assyrian texts, the Medes were ruled by “city-lords” (Bel Ali). By the time the Medes were attested, they were already the largest group in the Zagros, including in the non-IE kingdoms of Mannaea and Ellipi, based on onomastics and toponymy of 1000-600 BCE “The Iranians were the largest group (maximum 45.37 – minimum 32.36%) in Greater Media” (Zadok, 2002). This is also considering that the Iranians were not averse to adopting non-IE names as is the case of Teispes and Kidin-Hutran, son of Kurlus.

 

Were the Medes and Persians Zoroastrian?

Recently there have been some theories that the religion of the Medes and Achaemenids is related to that of a pre-Zoroastrian Mazdaism. It has always been noted that not all the practices of the Medes and Persians do not resemble that of the Zoroastrian Sassanians. But why should Zoroastrianism be a stagnant religion for millennia, should there be no difference between 3rd-century Arianism and 21st-century Protestantism? After all, the “orthodox” Zoroastrianism promoted by the Sassanians is only one version with other forms such as the one that only accepts the Gathas being considered a heretical one in the Bundahishn. There is no attestation of this Mazdaism in any known Indo-Iranian group. From Ossetians, Scythians, and Vedic Aryans there is no evidence of an Ahura Mazda as a deity in any of these pantheons because he is an invention of Zarathustra based on an earlier *Dyeus Indo-Iranian god.

In the Behistun inscription, Darius claims that the Scythians and Elamite rebels did not worship Ahura Mazda. Ahura Mazda is only invoked in the Old Persian and Elamite versions of the Behistun inscription; otherwise, he is replaced by Bel. Despite all the recent ‘Darius the Liar’ theories, Darius seems to be rather reliable in his accounts of battles compared to the previous Near Eastern monarchs as he constantly gives credit to his generals so there has to be a reason he singled these two groups. The Elamite rebels, Açina son of Upadarma and Martiya, all have Iranian names making it very likely that Elam was considered part of the Iranian world at the time of Darius. Considering that the Scythians also belong to the Iranian world, it seems that the worship of Ahura Mazda was confined to ethnic Iranians.

The idea that the Medes or Achaemenids adopted Zoroastrianism in the 7th-5th centuries BCE has little support. Already in the Achaemenid period, Zoroaster was associated with Media. The religion of the Medes was not much different from the Younger Avestan one as names such as Mitraku and Mazdaku are attested in the Neo-Assyrian period. It is noted that the later East Iranian languages are not directly related to Old Avestan, “While its geographical horizon is that of Southern Central Asia and Eastern Iran, it displays no phonological features characteristic of Eastern Iranian languages of later periods (Sims‐Williams 1998: 136)” (Hintze, 2015), the later East Iranian deities as well, eg Vakhshu and Sasana, do not appear in the Avesta. The current version of the Avesta is passed down from the Achaemenids, who replaced Apam Napat with their favorite goddess Anahita.

 

A Few Unknowns

1. While we do not have IVCp samples earlier than 3100BCE, the proposed date for the formation of the IVCp is 4300-3800 BCE and I propose an entrance for Proto-Indo-Aryan with the destruction of the Kot Diji Phase. I believe we’ll see some sort of local shifts in the Indus around that time. Indeed, from the Archaeological supplement A to Damgaard et al. 2018: “Fourth, if there was any Central Asian influence on South Asian populations, that influence likely long predated any development of Iranian, let alone Indo-Aryan, languages, and most likely occurred during the late NMG IV to early NMG V period (ca. 2800–2300 BCE)”, of course, their claims when it comes to the age of I-Ir should be ignored.
2. Since there is simply no consensus amongst linguists on who the supposed non-IE people of the Rigveda were, whether they were one group or not. If they were different groups then they could be from anywhere from Kelteminar-related groups, the previous/non-IE inhabitants of the Indus region, from beyond the Oxus, foreign traders from across the sea..etc. If they were related to Elamite as Blazek claims then there are only two likely scenarios:
   a. from when the “Proto-Elamite” network reached SiS around 3000BCE, in this case their presence in the Rigveda should be seen as ahistorical
   b. If there was an actual Elamite-related presence in Jiroft, seeing as the Elamite culture was at its peak at the time of the Shimashki Empire it seems plausible, and with Jiroft being the enemies of BMAC it could be possible that some of the parts of the Rigveda were based on historical events.
3. The linguistic model for PIE. While there are many possible alternative linguistic models such as a wave type of model or PIE being a creole language this post is based on a simplistic Eric Hamp model which is gaining more mainstream support recently although still confined to the Steppe model which makes it impossible unless they consider Afanasievo Indo-Iranian, the other west Asian PIE models are late neolithic Zagros or Armenia which does not fit with the copper age time frame. In my view, PIE is from the Alborz Piedmont, 5200-4700 BCE at the earliest, which allows for a movement west to the Caucasus and Anatolia and east to Turan before the region was abandoned by 3000 BCE.

 

Summary

Indo-Iranian has to be an early split as the Hamp model proposes. The most likely scenario for the development of Proto-Indo-Iranian is the beginning of a hierarchical society in northeastern Iran after 3400BCE and its spread to “Turan” forming the Namazga III culture (3200 BCE) during what seems to be a violent period. This Proto-Indo-Iranian Namazga III culture expanded all the way south to Shahr-e Sukhteh, Mundigak, and Mehrgarh. This is followed by a regionalization period (2800-2600 BCE) and by the end of it Indo-Iranian should have split 2500 BCE at the latest. Indo-Aryan languages should have entered India around that period by the end of the Kot Diji phase. Indo-Aryan languages would diversify in that region and Sanskrit and the Indra cult spread amongst the elites first with most Indo-Aryan languages that are spoken today descend from different dialects. Iranian most likely is the older split as proposed by Hamp as it is the more culturally and religiously diverse group. Already in the late Rigveda Iranian names are pretty diverse with some pretty similar to a Younger form of Avestan while others closer to a form of West Iranian.

The chronology of the religious texts should be Rigveda from 1900 to 1700 BCE (if the Anukramanis is right, which most definitely it’s not so, probably even as late as 1500 BCE) > Older layers of the Yashts > Atharvaveda with the Gathas either contemporary with the early Rigveda or slightly earlier. There must have been a gap between both Gathas, which was most likely composed in Eastern Parthia/Western Aria, and the Yashts as evidenced by the change in belief and language, which also predates the Iranian migration to the west. Despite the similarity in the language, due to the later sound redactions as both books were being compiled, the same is true for the Rigveda and the Atharvaveda where the first represents a “higher culture” and a different form of rituals while the latter matches with the cultural decline seen in the Iron Age.

The Mittani-related Aryans must have reached western Iran around 1800 BCE at the latest and are pretty much a movement of the Indra worshippers of BMAC as seen by the material culture brought by them, from the three-headed dragons to the Garuda birdman, as well as BMAC ancestry in some sites in the western Asia. Whether they were actual Indo-Aryan elites or the Indo-Aryanized Iranians or whether they brought Iranians along or was the Iranian in Issuwa from a separate movement is difficult to say. The main West Iranian migration should have been the typical proposed one between 1500-1000 BCE.

 

References

Anthony, David W. 2007. The Horse, the Wheel, and Language: How Bronze-Age Riders from the Eurasian Steppes Shaped the Modern World.
Batty, J. 2007. Rome and the Nomads: The Pontic-Danubian Realm in Antiquity.
Blažek, Václav. 2002. Elamo-Arica.
Bonora, Gian Luca. 2006. Archaeological Perspectives in the Linguistic Reconstruction of Protohistorical Baluchistan.
Bonora, Gian Luca. 2020. The Oxus Civilization and the Northern Steppes.
Bonora, Gian Luca, and Maurizio Vidale. 2013. “The Middle Chalcolithic in Southern Turkmenistan and the Archaeological Record of Ilgynly-Depe.”
Bryant, Edwin F. 2001. The Quest for the Origins of Vedic Culture: The Indo-Aryan Migration Debate.
Burrow, Thomas. 1973. “The Proto-Indoaryans.”
Crouwel, Joost H. 1979. Chariots and Other Means of Land Transport in Bronze and Early Iron Age Greece.
Diakonoff, Igor M. 1996. “Pre-Median Indo-Iranian Tribes in Northern Iran?”
Dubova, N., and B. Lyonnet. 2020. The World of the Oxus Civilization.
Eidem, Jesper. 2014. “The Kingdom of Samsi-Adad and Its Legacies.”
Eidem, Jesper, and Jørgen Læssøe. 2001. The Shemshāra Archives.
Erdósy, George. 1995. The Indo-Aryans of Ancient South Asia: Language, Material Culture, and Ethnicity.
Francfort, Henri-Paul. 2008. “A Note on the Hasanlu Bowl as Structural Network: Mitanni-Arya and Hurrian?”
Henkelman, W. F. H. 2011. “Cyrus the Persian and Darius the Elamite: A Case of Mistaken Identity.”
Hintze, Almut. 2015. “Zarathustra’s Time and Homeland: Linguistic Perspectives.”
Kenoyer, J. M. 1995. “Interaction Systems, Specialized Crafts, and Culture Change: The Indus Valley Tradition and the Indo-Gangetic Tradition in South Asia.”
Kenoyer, J. M. 2004. “Culture Change During the Late Harappan Period at Harappa: New Insights on Vedic Aryan Issues.”
Kershaw, P. K. 2000. The One-Eyed God: Odin and the (Indo-) Germanic Männerbünde.
Kosyan, Aram. 2006. “An Aryan in Isuwa, Iran, and the Caucasus.”
Kuz’mina, E. E. 1983. “Stages in the Development of Wheeled Transport in Central Asia During the Aeneolithic and Bronze Ages (On the Problem of the Migration of Indo-Iranian Tribes).
Lamberg-Karlovsky, C. C. 1967. “Archeology and Metallurgical Technology in Prehistoric Afghanistan, India, and Pakistan.”
Lamberg-Karlovsky, C. C. 2002. “Archaeology and Language: The Case of the Bronze Age Indo-Iranians.”
Lebedev, A. V. 2021. “Indo-Aryan Names in the Saga of Argonauts, Onomastics of Colchis, and Greek Inscriptions of the Northern Black Sea Region.”
Lincoln, Bruce. 1979. “The Indo-European Myth of Creation.”
Mallory, J. P., and D. Q. Adams. 1998. Encyclopedia of Indo-European Culture.
Masson, V. M. 1992. The Bronze Age in Khorasan and Transoxania.
Mousavi, Ali. 2008. “Late Bronze Age in North-Eastern Iran: An Alternative Approach to Persisting Problems.”
Olbrycht, Marek J. 2000. “Remarks on the Presence of Iranian Peoples in Europe and Their Asiatic Relations.”
Parpola, Asko. 1995. “The Problem of the Aryans and the Soma: Textual-Linguistic and Archaeological Evidence.”
Parpola, Asko. 2015. The Roots of Hinduism: The Early Aryans and the Indus Civilization.
Thompson, George. 2002. “Adhrigu and Drigu: On the Semantics of an Old Indo-Iranian Word.”
Von Dassow, Eva. 2014. “Levantine Polities Under Mittanian Hegemony.”
Watkins, Calvert. 2004. “The Third Donkey: Origin Legends and Some Hidden Indo-European Themes.”
Wilhelm, Gernot. 1989. The Hurrians.
Wilkinson, T. C. 2014. Tying the Threads of Eurasia: Trans-regional Routes and Material Flows in Transcaucasia, Eastern Anatolia, and Western Central Asia, c. 3000-1500 BC.
Winkelmann, Sylvia. 2020. “BMAC Glyptics: Typology, Context, Function, and Background.”
Witzel, Michael. 1998. “The Home of the Aryans.”
Witzel, Michael. 1999. “Early Sources for South Asian Substrate Languages.”
Witzel, Michael. 2000. “The Languages of Harappa: Early Linguistic Data and the Indus Civilization.”
Witzel, Michael. 2003. “Linguistic Evidence for Cultural Exchange in Prehistoric Western Central Asia.”
Witzel, Michael. 2004. “F.B.J. Kuiper (1907–2003).”
Young, T. Cuyler. 1976. “The Iranian Migration into the Zagros.”