Wednesday 16 August 2023

A reappraisal of the Indo-European migration from Iran starting from the new linguistic paper of Heggarty

Nine years ago I proposed a west Iranian homeland (in the Zagros mountains) for Indo-Europeans (see here). After that, there were genetic studies that discovered a Caucasus Hunter Gatherers component that spread to the steppe and a similar Iranian Neolithic component (in the Zagros) that spread to the east, and when it was discovered that this component was present in Bronze Age Anatolia without steppe ancestry, the South of Caucasus homeland has received new support. Now, also in linguistics, there is a trend in that direction, and a new, big stone has been thrown in the pool of Indo-European studies:

Paul Heggarty et al. , Language trees with sampled ancestors support a hybrid model for the origin of Indo-European languages.Science381,eabg0818(2023).DOI:10.1126/science.abg0818

It is a linguistic study based on lexical evolution, comparing the vocabulary of 161 Indo-European languages, and connecting the results with geography and genetics. The map above shows the general reconstruction: Proto-Indo-European started to split around 8100 years ago, south of Caucasus, in the region of origin of Caucasus Hunter Gatherers/Iran Neolithic ancestry. As they write: "This CHG/Iranian component is found first south of the Caucasus, including in the north to northeastern arc of the Fertile Crescent, among early farmers on the flanks of the Zagros Mountains in western Iran. The same CHG/Iranian ancestry component also admixes heavily (by ~5000 yr B.P.) into the region where languages of the Anatolian branch are first documented. CHG/Iranian is the dominant ancestry in ancient Armenia and Iran, in BMAC, and in most present-day populations who speak languages of the Iranic branch. It is also a major ancestry component among speakers of the Indic branch, particularly in regions furthest from the Dravidian-speaking (i.e., non–Indo-European) south of India. Thus, it is the CHG/Iranian ancestry component that most strongly connects the past populations who potentially spoke the branches of Indo-European in Europe and south (and east) of the Caucasus. [...] we propose a hybrid hypothesis in which Indo-European languages spread out of an initial homeland south of the Caucasus, in the northern Fertile Crescent. Only one major branch spread northward onto the steppe and then across much of Europe."
This major branch would be the NW Indo-European group, already known in Indo-European studies, including Italic, Celtic, Germanic, Balto-Slavic, but without Tocharian that is usually included in the group. Archaeologically and genetically, it seems that all this group is originally connected with the Corded Ware culture, from which derived also the central European Bell Beaker people. The separation of this big branch is here dated 6981 BP (5645‒8395), around the time of the arrival of CHG in the steppe according to ancient DNA studies.  

Albanian, Armenian and Greek instead are placed apart (and not with a SE group with Indo-Iranic as proposed by some), and separated around 6135 BP (4540‒7882). So, as shown in the map, they would have migrated through Anatolia to the Balkans without passing by the steppe. This is denied by the paper of Lazaridis, "The genetic history of the Southern Arc" that proposed instead that Greeks and Armenians came from Yamnaya expansions, bringing steppe ancestry to Greece and Armenia during the Bronze Age. This view is also consistent with the kind of R1b Y DNA lineages (L23) that is shared by ancient Greeks and Armenians in the 2nd millennium BCE, connected with Yamnaya.  
Moreover, the separation of Anatolian is dated 6932 BP (5403‒8613), which means that Greeks moved to Anatolia later than Anatolians, and it is not clear how this could happen and why they did not settle somewhere in Anatolia and did not push Anatolians towards Greece. One can observe that in Greece there are some famous Anatolian-like toponyms like Parnassos. But I have to cite here from a work of Tardivo and Kitselis: "the mountain name Παρνασσός has a Luwian equivalent Parnassa, which is believed to be derived from the Luwian root parna ‘house’, a word that lacks secure cognates outside Anatolia. Comparable material is found nearby in the Mediterranean region like the Egyptian pr “house” and Hurrian pur(u)li “house”."
Anyway, this can be explained as the substrate of Anatolian migrants with a CHG ancestry who arrived in Greece already at the beginning of the Bronze Age.

About Indo-Iranic, their separation is dated 6981 BP like the NW branch, as if they departed at the same time. The separation of Indic, instead, is dated 5520 BP (4535‒6796), as if they reached South Asia much later, as is suggested by the map indeed. I appreciate very much that they accept a route directly through Iran rather than through the steppes, but I find the movement towards India too slow.
Now, although dates of known historical language changes correspond quite well to the dates of this study based on lexical comparison, I have many doubts on the validity of these prehistoric dates, but let us see if there is some correspondence with what we know from archaeology.

The initial division of Indo-Europeans is so described: "Our hybrid hypothesis posits that out of this homeland south of the Caucasus, from ~8120 yr B.P., PIE began to diverge as early migrations split it into multiple early branches." Now, 8120 BP means 6170 BCE. It is an interesting period, because it is associated with a dramatic climatic change, the so-called 8.2-kiloyear event. This event, apparently starting from the Atlantic, "like the Younger Dryas and 9.2ka BP event, was marked by abrupt cooling and aridity across much of the world" (Matthews and Nashli 2022). According to an article by Flohr et al. of 2015 ("Evidence of resilience to past climate change in Southwest Asia: Early farming communities and the 9.2 and 8.2 ka events"), "The so-called 8.2 ka event was one of the most pronounced and abrupt Holocene cold and arid events. [...] Sharp decreases in air temperature of up to 4°C are also evident in many other high-resolution proxy records from the circum-Atlantic and Mediterranean basin. [...] It appears that wetness increased north of 42° latitude, while aridity increased south of this. For the Eastern Mediterranean, a reduction in precipitation of around 17% has been calculated. The event is well-dated to have taken place between 8250 and 8000 cal BP and to have lasted around 160 years." About the consequences, Matthews and Nashli write: "it has been suggested that the Neolithic movement into the Zagros foothills and Fars may have been stimulated by this century-long episode. [...] across the plains of central and northern Iran there is as yet no convincing sign of local precursors to the well-adapted Neolithic farmers and herders who spread into the region from c. 6000 BC, bringing their domesticated herds and grains with them. Where did they come from? The likeliest hypothesis for the time-being is that they moved into central and northern Iran from the west, steadily advancing across the Zagros ranges and foothills from areas where farming had already been practiced for up to 2000 years. Why did they move? To what extent their movement was stimulated by climatic adversity attendant upon the 8.2 ka BP event remains unclear, but there appears to be an at least approximate contemporaneity. [...] As to the dispersal of Neolithic lifeways eastwards from Iran into Turkmenistan and beyond, the many material culture connections between the Neolithic sites of north-eastern Iran and those of southern Turkmenistan across the Kopet Dagh, such as Monjukli Depe, are highly suggestive. The first point to make is that, yet again, the appearance of the earliest Jeitun culture sites in Turkmenistan appears to coincide with the 8.2 ka BP event (Harris 2010: 233; see also Düring 2011: 124–125 for possible impact of the 8.2 ka BP event on agricultural expansion across Turkey). The presence of domesticated sheep by 6000 BC at Obishir V in southern Kyrgyzstan suggests a relatively rapid spread of Neolithic herding practices eastwards from southern Turkmenistan across challenging terrain. Secondly, there is no evidence for local development of the Neolithic in Turkmenistan or beyond. David Harris summarises the Neolithic of southern Turkmenistan thus: 
The general uniformity of the material culture of the Jeitun-Culture settlements, especially their mudbrick architecture and chaff-tempered pottery, supports the inference that they were initially founded as sedentary settlements by migrants seeking new land to occupy with their crops and livestock. (Harris 2010: 233) 
Finally, Harris concludes that the dispersal of Neolithic settlers, with their herds and plants, towards Turkmenistan took place across the northern reaches of Iran, an “inviting corridor” leading from the Zagros to the Kopet Dagh."

So, possibly before 6000 BCE the sudden cold and aridity pushed people down from the north-central Zagros towards the Caspian region and beyond, to Turkmenistan, areas where we later find the cradle of the Iranic world. A Neolithic presence in the Caspian region was already around 7000 BCE in Sang-e Chakhmaq West, but it was much more limited, and did not reach Central Asia. Interestingly, also in Baluchistan we have a new kind of Neolithic dated from 6000 BCE by Jarrige (although Ceccarelli and Petrie give a date of c.5470-4700 BCE from the deposits of Period IIA), in Mehrgarh Period II: "Pottery appeared in Period II; it is vegetal-tempered, comparable to the Neolithic pottery found in Iran including at the site of Tepe Yahya and in the Daulatabad Plain. Some Period II vessels display applied decorations, a feature observable on vegetal-tempered vessels from Tal-i Iblis. Jean-François Jarrige, however, has compared these decorations more specifically with those from Umm Dabaghiyah in northern Iraq. Additionally, the technique used to create this pottery, namely sequential slab construction, has also been identified on Neolithic vessels from Iran as well as at Umm Dabaghiyah." (Mutin and Garazhian 2021) Jarrige (2008: 150) made also another comparison with Umm Dabaghiyah about Period II: "The impressive plans of compartmented buildings of Period IIA can be compared with buildings with similar plans from Mesopotamian sites such as Tell el Oueili or Umm Dabaghiyah at the end of the 7th millennium BC. It is probably not a mere chance if one notices the occurrence at Umm Dabaghiyah and at Mehrgarh, Period IIA, of some potsherds not only built according to the same sequential slab construction but also bearing similar applied designs." 
Now, a connection of Umm Dabaghiyah in North Mesopotamia with Mehrgarh in Baluchistan seems quite unlikely at first sight. But there are also connections with an important site southwest of the Urmia lake, Hajji Firuz Tepe in Iranian Azerbaijan, as Matthews and Nashli report: "Concerning the origin of the first settlers at Hajji Firuz, on the basis of the pottery Voigt (1983: 166) argues for connections across the northern Zagros to the west into Upper Mesopotamia, with sites such as Telul eth-Thalathat, Sotto and Umm Dabaghiyah." Voigt (see here, p.166) spoke of a single tradition of manufacture and decoration of ceramics in Hajji Firuz and early Hassuna, which means Umm Dabaghiyah. She proposed that the area (Ushnu-Solduz valley south of Lake Urmia) was settled by cultivators and herders coming from the west, where they were in contact with early Hassuna groups or with their ancestors, and found a continuity of contacts in the later period. She also wrote (p.168) of a chain of sites from Hassuna-Samarra in central Mesopotamia to Tepe Sarab in Mahidasht, to Tepe Sialk in the central Iranian plateau. To extend the chain up to Mehrgarh, we can add what Fagan (1996: 437) has written: "At Mehrgarh, continuity and change is marked with the introduction of soft ware, Buff Ware or Chaff Tempered Pottery in period Mehrgarh IIA. This ceramic seems to have broad similarities on the Iranian plateau (e.g., Yahya period vii-v, Tepe Sialk, Belt & Hotu Caves) perhaps far west in the Zagros (Jarmo)".  We can also add a direct parallel by Vandiver (1987: 18) between the ceramic of Hajji Firuz and Mehrgarh: "At Hajji Firuz about twenty sherds with negative basket impressions in the base have delaminated from an outer surface layer which has a positive but less distinct basket impression. At Mehrgarh, J.-F. Jarrige has reported about five similar sherds, in addition to some with basket straw still embedded in the wall." And Ceccarelli-Petrie (2020: 4): "what is referred to as Burj Basket-Marked ware was among the earliest ceramics identified in the archaeological record of South Asia. For this technique, potters used baskets as a mould, and vessels were often coated with a clay slip to hide basket impressions. This combination of techniques also seems to have been quite widespread from the Near East to South Asia, as suggested by evidence from Abu Hureyra (Syria), Ali Kosh and Hajji Firuz Tepe (Iran)." About Ali Kosh at the southwestern border of Central Zagros (dated from 7500 to 6500 BCE), Jarrige (2008: 151-152) noticed other affinities already from Period I, that he dates from before 7000 BCE (while Petrie from 6000 BCE): "the full setting of farming economy at Mehrgarh displays evident similarities with what had been noticed in the case of the early Neolithic settlements in the hilly regions forming the eastern border of Mesopotamia. The circular houses of the earliest Neolithic villages have not been found at Mehrgarh. But quadrangular houses built with about 60 cm long narrow bricks with a herringbone pattern of impressions of thumbs to provide a keying for the mud-mortar, have been uncovered at several aceramic Neolithic sites in the Zagros, such as Ganj Dareh or Ali Kosh in the Deh Luran region of Iran, where, like at Mehrgarh, traces of red paint have also been noticed on the walls. Circular fire-pits filled with burnt pebbles are also associated to all these early settlements. The lithic industries also show evident parallels [...] polished-stone axes begin occurring at several sites of the Deh Luran area, such as Ali Kosh, only in the later phases of the aceramic Neolithic along with an increasing number of stone vessels. It is the same at Mehrgarh where the polished stone axes in black diorite are found only in the upper levels of Period I, mostly as gravegoods. [...] the few graves exposed at Ali Kosh show skeletons with positions rather similar to those of Mehrgarh. Among the gravegoods one notices ornaments made of seashells and semi-precious stones such as turquoise, a few beads in copper. Baskets coated with bitumen and oblong-shaped cakes of red-ochre strengthen the parallels."

The subsequent innovations appearing both in sites between Mesopotamia and central Zagros and in Mehrgarh suggest a repeated contact between these distant regions, as if it followed an established route. In Sotto, a site near Umm Dabaghiyah, there is also the earliest finding of lapis lazuli in Mesopotamia, which would confirm exchanges with Central Asia (see here). But then, do we have just a chain of cultural exchanges from Mesopotamia to Baluchistan, or also a movement of people? And did this movement start from Mesopotamia? In a passage about the site of Yanik Tepe (northeast of Lake Urmia) Matthews and Nashli write: "Objects include many alabaster bowl and bracelet fragments, comparable to those from Neolithic sites of the Zagros, and bone and obsidian tools using obsidian from south Caucasian and north Iranian sources. A small stone figurine takes the form of a human head with clear representation of artificial cranial elongation (Figure 5.45), also a key trait of the Zagros Neolithic. Chaff-tempered pottery, occasionally painted, compares well to Neolithic ceramics from Hajji Firuz to the southwest of Lake Urmia. The alabaster bracelets and evidence for cranial elongation suggest a Zagros origin for the Neolithic settlers of the Lake Urmia basin rather than a development from local hunter–gatherer communities, until now conspicuous by their absence from the archaeological record." 

So, did people of Lake Urmia in Late Neolithic come from the central Zagros or from Mesopotamia?
Ajorloo (2008: 114) observes that there are two Pottery Neolithic sites, Ahrendjān and Qara Tepe to the northwest of Lake Urmia, earlier than Hajji Firuz: "According to the field data it is possible to prove that the Hajji Firuz is not the first stage in the formation process of the early villages in Azerbaijan. Therefore it seems the hypothesis of the migrant people of the Hassuna sphere to the west bank of the Urmia Lake can not explain the Neolithization process in the region." And in a later paper (Ajorloo 2016: 157): "while Ahrendjān-Qara Tepe wares stand in a range attested to the Proto-Hassuna horizon, c. 6700-6200 BC, the radiocarbon dates put Hājji Firūz not earlier than c. 6000 BC, and, moreover, there are some remarkable resemblances between the Ahrendjān-Qara Tepe ceramic assemblage and Sarāb, Gūrān II, Abdul Hosein II and even Jarmo II which fall within a range of c. 7000-6500 BC. Such a chronology is also roughly in accordance with the retreat of Lake Urmia and the emergence of a transhumance way of life in the Central Zagros. Consequently, in general terms, it is reasonable to propose a relative dating for Ahrendjān-Qara Tepe as c. 6500-6000 BC." But if the Urmia basin was settled by transhumant herders from the central Zagros, this does not exclude that also people from Mesopotamia arrived, probably agriculturists, especially in Hajji Firuz, where the connections with Hassuna culture are particularly strong: "In addition to incised decoration, like the Hassuna pottery, the pottery tradition of Hājji Firūz includes husking trays, pithoi and wide-mouth wares; such types being related to agricultural activities. On the contrary, in the Ahrendjān-Qara Tepe tradition they had been using forms appropriate to dairy products like spouted vessels, collared jars and narrow mouth vessels. The husking tray is recognized as one of the indicators for the Late Neolithic period in northern Mesopotamia and the Hassunan/Samarran horizon but as yet no husking tray has been recorded from the Ahrendjān-Qara Tepe tradition and the well-known Neolithic sites in the Zagros, for example Jarmo II, Gūrān II, Abdul Hosein II, Sarāb and Ganj Dareh A-D" (Ajorloo 2016: 153).

Today luckily we have a genetic study of Hajji Firuz around 6000 BCE or slightly later, that has revealed that the ancestry of the people was double: 53% Anatolia Neolithic, 47% Ganj Dareh Neolithic (Narasimhan 2019, supplement, p.196). So, the answer to our question seems to be that the people of Hajji Firuz came from both directions. Not only, we also have the Y DNA haplogroups: two of them are J2b, a haplogroup that was found also in Tepe Abdul Hosein around 8000 BCE, in central Zagros Neolithic, with a pure Iran Neolithic autosomal ancestry. This means that J2b was part of Iran Neolithic people of central Zagros, and interestingly it is quite frequent in Albanians, Greeks, and South Asians. J2b2 was found also in three Mycenaean samples (Mygdalia) and in several Balkanic samples of the Bronze Age, probably belonging to proto-Illyrians (see here). 
About the autosomal ancestry, the study of Broushaki on the three samples from Abdul Hosein and one from Wezmeh Cave (a later site in central Zagros) found out that "Zoroastrians are the most genetically similar to all four Neolithic Iranians, followed by other modern Iranians (Fars), Balochi (SE-Iran, Pakistan and Afghanistan), Brahui (Pakistan and Afghanistan), Kalash (Pakistan) and Georgians". Most of these peoples are Indo-Iranian speakers, Dravidian Brahuis are genetically very similar to Baluchis, and Georgians are geographically close and have special connections with Indo-European.
Moreover, according to Ajorloo (2016: 153), among all central Zagros sites, Abdul Hosein II shows more similarities in term of shapes and forms of the ceramics with Ahrendjān and Qara Tepe near Lake Urmia: we can suppose that some people from this village went to the north in search of pastures and their descendants settled also in Hajji Firuz. 
On the maternal side, one of the two men with J2b from Hajji Firuz Tepe had mtDNA K1a3, and K1a has been found not only in Barcın in Northwestern Anatolia Neolithic, but also in two Pre-Pottery Neolithic sites in Mesopotamia, Boncuklu Tarla near Mardin and Çayönü (where also specifically K1a3 is present), and also in Proto-Neolithic Shanidar, on the Iraqi side of the Zagros. These Mesopotamian and Zagros sites have a significant amount of Central Zagros Neolithic ancestry, and Çayönü has also revealed a female outlier with more of that ancestry and also with artificial cranial deformation, a feature (caused by head bandage) well known in central Zagros (Ganj Dareh, Abdul Hosein), in Shanidar Cave, Ali Kosh, and also, as we have seen, in the figurines of Yanik Tepe in the Urmia region. The Çayönü male with K1a3 mtDNA has J2a1a as Y DNA, which connects him with Caucasus Hunter Gatherers and Iran Mesolithic: J2a has been found in Kotias Klde in Georgia (8th mill. BCE), in Hotu Cave south of Caspian Sea (dated 9100-8600 BCE) and in Neolithic Tepe Guran in central Zagros (6700 BCE). Moreover, J2a1a, besides later Anatolian samples (including Phrygian Gordion), was also in Armenia Neolithic (Aknashen and Masis Blur), in Mycenaean Greece, in Tepe Hissar (3640-3518 calBCE), in Geoksyur in Turkmenistan (3365-3097 calBCE), in Shahr-i Sokhta (2600-2500 BCE), and in the Urmia region in the Late Bronze/Iron age (Dinkha and Hasanlu Tepe). And J2a1h was the important Indus_Periphery_West individual from Shahr-i Sokhta (and another Indus Periphery sample there is J2a), but also a local individual from the same site, one of Tepe Hissar (3641-3519 calBCE), and some later individuals from Swat and Central Asia. J2a1h is also especially frequent among Zoroastrians in Iran (15.4% in Tehran and 17.6% in Yazd, and also 17% of Persians from Yazd, as shown by Grugni 2012, where it is called J2a3h, while J2a3*-Page55=J2a1* is even 23% in Zoroastrians from Tehran).
Thus, J2a1 has a good connection with Indo-Iranians, Greeks, Phrygians and Anatolians, very conservative Indo-European branches for many morphological and phonetical aspects, and not sharing the special vocabulary characteristic of the NW Indo-European branch. And clearly this haplogroup starts from south of the Caucasus. 

The Indus_Periphery_West individual from Shahr-i Sokhta, dated 3100-3000 BCE, is important because it was analyzed in the paper of Shinde et al. (2019: 4): "To obtain insight into the origin of the Iranian-related ancestry in the IVC Cline, we co-modeled the highest-coverage individual from the IVC Cline, Indus_Periphery_West (who also happens to have one of the highest proportions of Iranian-related ancestry) with other ancient individuals from across the Iranian plateau representing early hunter-gatherer and food-producing groups: a 10,000 BCE individual from Belt Cave in the Alborsz Mountains, a pool of ⁓8000 BCE early goat herders from Ganj Dareh in the Zagros Mountains, a pool of ⁓6000 BCE farmers from Hajji Firuz in the Zagros Mountains, and a pool of ⁓4000 BCE farmers from Tepe Hissar in Central Iran. [...] The only consistently fitting models specified that the Iranian-related lineage contributing to the IVC Cline split from the Iranian-related lineages sampled from ancient genomes of the Iranian plateau before the latter separated from each other [...] The Belt Cave individual dates to ⁓10,000 BCE, definitively before the advent of farming anywhere in Iran, which implies that the split leading to the Iranian-related component in the IVC Cline predates the advent of farming there as well. Even if we do not consider the results from the low-coverage Belt Cave individual, our analysis shows that the Iranian-related lineage present in the IVC Cline individuals split before the date of the ⁓8000 BCE Ganj Dareh individuals, who lived in the Zagros mountains of the Iranian plateau before crop farming began there around ⁓7000–6000 BCE. Thus, the Iranian-related ancestry in the IVC Cline descends from a different group of hunter-gatherers from the ancestors of the earliest known farmers or herders in the western Iranian plateau." 
What they say here about Ganj Dareh is not correct, because the Neolithic people of all levels (8200-7600 BCE) there used domesticated cereals: "Regarding plant exploitation, the people of Ganj Dareh utilised domesticated two-row hulled barley (Hordeum distichum) in increasing amounts through levels E to A" (Matthews and Nashli 2022: 69). But also the analysis and the conclusions were contested by another paper, by Maier et al. (2023: 50): "The combined Indus Periphery group we analyzed included seven individuals from Shahr-i-Sokhta and three individuals from Gonur (three individuals were removed from the Narasimhan et al., 2019 dataset due to potential contamination with modern human DNA and low coverage). We removed one individual from the Ganj Dareh Neolithic group as potentially contaminated, and one second- or third-degree relative was removed from the Anatolia Neolithic group [...] A model ‘Indus Periphery = Ganj Dareh Neolithic + Onge (ASI)’ was strongly rejected for the Indus Periphery group of 10 individuals [...] and a model that was shown to be fitting for all Indus Periphery individuals modeled one by one by Narasimhan et al. (Ganj Dareh Neolithic + Onge (ASI) + West Siberian hunter–gatherers (WSHG)) was rejected for the grouped individuals [...] In contrast, a model ‘Indus Periphery = Ganj Dareh Neolithic + Onge (ASI) + WSHG + Anatolia Neolithic’ was not rejected [...] and produced plausible admixture proportions for all four sources that are confidently above zero: 53.2 ± 5.3%, 28.7 ± 2.1%, 10.5 ± 1.3%, and 7.7 ± 2.9%, respectively" 

This means that an Anatolia Neolithic component, clearly present in Hajji Firuz, Tepe Hissar and other post-6000 BCE Iranian sites, should be recognized also in Indus Periphery individuals, and so a migration from western Iran is not rejected. They also add (p.51): "we show four graphs with four admixture events that model the Indus Periphery group as a mixture of three or four sources, with a significant fraction of its ancestry derived from the Hajji Firuz Neolithic or Tepe Hissar Chalcolithic lineages including both Iranian and Anatolian ancestries. [...] a key historical conclusion of the study (that the predominant genetic component in the Indus Periphery lineage diverged from the Iranian clade prior to the date of the Ganj Dareh Neolithic group at ca. 10 kya and thus prior to the arrival of West Asian crops and Anatolian genetics in Iran) depends on the parsimony assumption, but the preference for three admixture events instead of four is hard to justify based on archaeological or other arguments. [...] the inconsistency reflects the fact that the deeply diverging WSHG-related ancestry (Narasimhan et  al., 2019) present in the IVC (Indus Valley Civilization genetic grouping, which is the same group as Indus Periphery) at a level of ca. 10% was not taken into account explicitly" 
See here for the 4 graphs. Here are two examples:

As we see, it is possible that Indus Periphery received admixture from an ancestry close to Tepe Hissar, or from an ancestry related to Hajji Firuz and another Iranian source parallel to Ganj Dareh. Many possibilities are there, but considering Y DNA, the fact that J2a1h is found also earlier in Tepe Hissar can be a hint of a connection. It can be also significant that according to the skeletal studies of Hemphill, Lukacs and Kennedy, Harappans had the closest affinities with Tepe Hissar period III. 
An interesting detail about the Indus Periphery West individual is that in his burial "Salvatori et al. note that a distinctive pottery type among the grave goods belongs to a distinctive cluster of graves at Shahr-i-Sokhta that is “possibly local or northeast oriented (Kandahar area)”" (Narasimhan 2019, supplement, p.115). Maybe this individual was from the region of Kandahar, where there was Mundigak, a site that was possibly founded by people coming from Baluchistan according to Jarrige (1993: 26):

"Work conducted at Mehrgarh has clearly shown that the cultural assemblage of the preurban phases of Mundigak (period IV) is closely linked to Baluchistan. The foundation of Mundigak can even be interpreted as the settling of people from Baluchistan who were probably aware of the importance of such a location for the control of the nearby mineral resources. The remains from period I at Mundigak fit almost perfectly the cultural assemblage of period III at Mehrgarh, dated to the end of the fifth and the very beginning of the fourth millennium b.c." 
That region, Greek Arachosia, had the Avestan Harahvaiti river, a name that corresponds to Sanskrit Sarasvati and has no other parallels in Iranic. We can suppose that people from Mundigak were Indo-Aryan speakers with an Indus Periphery genetic profile. About Mehrgarh III, the Chalcolithic phase, it has many developments compared with the Neolithic phase, but it is considered in continuity for architecture, ceramic technology, subsistence, although Hemphill, Lukacs and Kennedy noticed a great difference in the people of this phase compared to Period I, with close affinity with Iranian Tepe Hissar and Near Eastern skeletons, while Period I has affinity with later Inamgaon in Maharashtra, which suggests a type closer to South Indians. They placed the change thus between 6000 and 4500 BCE, but there is no analysis of Period II. In aceramic Period I there were already some elements from Iran/Mesopotamia, including kinds of wheat a probably also domesticated barley, although wild barley was also present. But we have seen that pottery and architecture of Period II had special connections with western cultural phenomena. Maybe, the initial aceramic Neolithic was mainly due to cultural contact, with local evolutions like zebu domestication, but in the course of time Iranian immigrants became dominant, especially after the 8.2 ka event. They colonized Baluchistan, bringing proto-Indo-Iranian language, during the 6th millennium BCE, progressively spreading to the east into South Asia, but also going to the west in Arachosia around 4000 BCE. Narasimhan et al. (2019: 11) suggest that the migration to South Asia is not later than 6000 BCE: "although our analysis supports the idea that eastward spread of Anatolian farmer–related ancestry was associated with the spread of farming to the Iranian plateau and Turan, our results do not support large-scale eastward movements of ancestry from western Asia into South Asia after ~6000 BCE (the time after which all ancient individuals from Iran in our data have substantial Anatolian farmer–related ancestry, in contrast to South Asians who have very little)" 
On the other hand, they place the admixture with Andamanese-like ancestry creating the Indus Periphery cline "by ~5400 to 3700 BCE", which is in the range of Period II-III of Mehrgarh in the chronology of Petrie. According to Heggarty, the separation of Indic from Indo-Iranic can be placed in 5520 BP (4535‒6796), that is, around 3570 BCE, with a large range starting from 4846 BCE. We can place the real development of Indo-Aryan after the assimilation of South Asian non-Indo-European people, with special features like retroflexion. The developments of Iranic can be due instead to the "Anatolian" influence, that possibly caused the loss of aspiration and spirantization.

But was there a memory in Iranic tradition of this migration? Avestan Vendidad 2 says (in the translation of Darmesteter):

5. And the fair Yima replied unto me, O Zarathushtra, saying: 'Yes! I will make thy world increase, I will make thy world grow. Yes! I will nourish, and rule, and watch over thy world. There shall be, while I am king, neither cold wind not hot wind, neither disease nor death.'
6. Then I, Ahura Mazda, brought two implements unto him: a golden seal and a poniard inlaid with gold. Behold, here Yima bears the royal sway!

8. Thus, under the sway of Yima, three hundred winters passed away, and the earth was replenished with flocks and herds, with men and dogs and birds and with red blazing fires, and there was room no more for flocks, herds, and men.
9. Then I warned the fair Yima, saying: 'O fair Yima, son of Vivanghat, the earth has become full of flocks and herds, of men and dogs and birds and of red blazing fires, and there is room no more for flocks, herds, and men.'
10. Then Yima stepped forward, in light, southwards, on the way of the sun, and (afterwards) he pressed the earth with the golden seal, and bored it with the poniard, speaking thus: 'O Spenta Armaiti, kindly open asunder and stretch thyself afar, to bear flocks and herds and men.'
11. And Yima made the earth grow larger by one-third than it was before, and there came flocks and herds and men, at their will and wish, as many as he wished.
12. Thus, under the sway of Yima, six hundred winters passed away, and the earth was replenished with flocks and herds, with men and dogs and birds and with red blazing fires, and there was room no more for flocks, herds, and men.
13. And I warned the fair Yima, saying: 'O fair Yima, son of Vivanghat, the earth has become full of flocks and herds, of men and dogs and birds and of red blazing fires, and there is room no more for flocks, herds, and men.'
14. Then Yima stepped forward, in light, southwards, on the way of the sun, and (afterwards) he pressed the earth with the golden seal, and bored it with the poniard, speaking thus: 'O Spenta Armaiti, kindly) open asunder and stretch thyself afar, to bear flocks and herds and men.'
15. And Yima made the earth grow larger by two-thirds than it was before, and there came flocks and herds and men, at their will and wish, as many as he wished.
16. Thus, under the sway of Yima, nine hundred winters passed away10, and the earth was replenished with flocks and herds, with men and dogs and birds and with red blazing fires, and there was room no more for flocks, herds, and men.

20. The Maker, Ahura Mazda, called together a meeting of the celestial Yazatas in the Airyana Vaejo of high renown, by the Vanguhi Daitya.
The fair Yima, the good shepherd, called together a meeting of the best of the mortals, in the Airyana Vaejo of high renown, by the Vanguhi Daitya.
21. To that meeting came Ahura Mazda, in the Airyana Vaejo of high renown, by the Vanguhi Daitya; he came together with the celestial Yazatas.
To that meeting came the fair Yima, the good shepherd, in the Airyana Vaejo of high renown, by the Vanguhi Daitya; he came together with the best of the mortals.
22. And Ahura Mazda spake unto Yima, saying: 'O fair Yima, son of Vivanghat! Upon the material world the evil winters are about to fall, that shall bring the fierce, deadly frost; upon the material world the evil winters are about to fall, that shall make snow-flakes fall thick, even an aredvi deep on the highest tops of mountains.
23. And the beasts that live in the wilderness, and those that live on the tops of the mountains, and those that live in the bosom of the dale shall take shelter in underground abodes."
And then, he had to create the Vara, a sort of underground garden where the best humans and animals and plants were preserved from the winter. Is this a memory of a special cold period like the 8.2 ka event? And where was really Airyanam Vaejo? Maybe we should reevaluate the identification of Bundahishn, that placed it "bordering Azerbaijan", and Darmesteter's identification of the river Daitya with the Araxes. Maybe we should think of the region around the Urmia lake, that has been a refuge also during the Bronze Age, when great part of Iranian sites were abandoned. And maybe we should include the central Zagros, the cradle of Iranian Neolithic, with its caves like Shanidar and Wezmeh. If there was a movement in the late Neolithic from there to the East Iranian and Central Asian Avestan homeland, did they completely forget their origin? 

Giacomo Benedetti, Impruneta, 17/8/2023



B. Ajorloo (2008) The neolithization process in Azerbaijan: An introduction to review, Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East Madrid, April 3-8 2006, Centro Superior de Estudios sobre el Oriente Próximo y Egipto, Madrid.

B. Ajorloo (2016) The Early Pottery Neolithic Tradition of the Salmās Plain in Azerbaijan, Northwestern Iranian Plateau, in "The Neolithic of the Iranian Plateau Recent Research" edited by Kourosh Roustaei & Marjan Mashkour, Ex Oriente, Berlin.

Alessandro Ceccarelli and Cameron Petrie (2020) Cultural Evolutionary Paradigms and Technological Transformations from the Neolithic up to the Indus Urban Period in South Asia, Apollo - University of Cambridge Repository.

B.M. Fagan (1996) The Oxford Companion of Archaeology, Oxford University Press, New York.

Jean-François Jarrige (1993) The Early Architectural Traditions of Greater Indus as Seen from Mehrgarh, Baluchistan, Studies in the History of Art , 1993, Vol. 31, Symposium Papers XV: Urban Form and Meaning in South Asia: The Shaping of Cities from Prehistoric to Precolonial Times, pp. 25-33.

Jean-François Jarrige (2008) Mehrgarh Neolithic, Pragdhara 18.

Robert Maier et al. (2023) On the limits of fitting complex models of population history to genetic data

Roger Matthews and Hassan Fazeli Nashli (2022) The Archaeology of Iran from the Palaeolithic to the Achaemenid Empire, Routledge, London and New York. 

Benjamin Mutin, Omran Garazhian (2021) Migrations, transfers, exchanges, convergences? Assessing similarities and differences among the earliest farmers between the Daulatabad and Kachi Plains (Southern Iran and Pakistan) Marc Lebeau. Identity, Diversity & Contacts. Proceedings of the International Congress The East 1, Brepols, pp.113-136, 2021, Identity, Diversity & Contacts. Proceedings of the International Congress The East 1, 978-2-503-58949-7. ffhal-03856097.

Narasimhan et al. (2019) The formation of human populations in South and Central Asia, Science 365.

Shinde et al. (2019) An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers, Cell, 

Sunday 15 May 2022

The Great Drought: going back on the chronology of ancient Indian dynasties

Vishvamitra looks at Rama who breaks the bow, winning the hand of Sita in marriage.

Around 12 years ago I wrote an article proposing a comparison between archaeology and Indian historical tradition, with some interesting results, see here. I followed there the genealogies of Pargiter (given in Ancient Indian Historical Tradition), but recently I have read a post by Benjamin Lloyd, on Viśvāmitra and his synchronisms, that place him at the time of Rāma Jāmadagnya or Paraśurāma but also of Rāma Dāśarathi, the king of Ayodhya, instead of 33 generations earlier, as in Pargiter's table. Lloyd's idea is that the Ikṣvāku genealogy, strangely longer than the other genealogies, is actually a mistaken collation of different and parallel genealogies into a single lineage, ignoring that some kings of the list are actually contemporary. So, around the same age can be placed not only Viśvāmitra and the two Rāmas, but also Triśaṅku, Divodāsa Pañcāla, his descendant Sudās, Arjuna Kārtavīrya and Māndhātṛ, as is clear also from another post of Lloyd, giving all synchronisms and tables. 
If we accept this theory, also figures present in Ṛgveda and other Vedic texts like Trasadasyu become part of a similar age (only 4 generations earlier), instead of being a different person as in Pargiter's theory, and the period of crisis and invasions of Arjuna Kārtavīrya is not different from that of Sudās Paijavana and his Battle of the Ten Kings. Not only, the mentions of the Great Drought of 12 years become almost completely related with the same period, that is given as the transition between Tretā and Dvāpara. Pargiter accepted (p.177 of AIHT) that Rāma Dāśarathi lived in that transition, but when the Mahābhārata (I.2, see here) places also Rāma Jāmadagnya in the same period, he had to deny it.

Why did I write 'almost completely'? Because in Nirukta (II.10) and Bṛhaddevatā (VII.155-VIII.2), a drought of 12 years is caused by the consecration as king of the Kuru Śantanu instead of his elder brother Devāpi, who, having a skin disease (according to the Bṛhaddevatā), decided to retire to practice austerities. The idea of the drought is certainly suggested by RV X.98, that is analyzed there, because that hymn is a prayer for rain, but there is no specific mention of the length of the drought. Also in the Viṣṇu Purāṇa (IV.20) it is said that in the kingdom of Śantanu there was no rain for twelve years, because the rightful king should have been the elder brother Devāpi, who retired to the woods, with no mention of skin disease. In Mahābhārata (V.147), instead, there is no mention of the drought, but only of the skin disease, that is presented as a right reason to exclude him from kingship. Now, Śantanu is clearly placed 4 generations before the Mahābhārata battle, at the end of Dvāpara Yuga and not at the transition between Tretā and Dvāpara. We can suppose that the idea of the 12 years' drought here has been adopted by the authors of the Nirukta and Bṛhaddevatā (and following that, of the Viṣṇu Purāṇa) because it was in the tradition, but in the Mahābhārata that drought is placed in a different period.
Let's see the various occurences of it.

1) Saṃvaraṇa. MBh I.163:
“And Saṃvaraṇa, that bull among men with due rites took Tapatī's hand on that mountain-breast which was resorted to by the celestials and the Gandharvas. The royal sage, with the permission of Vasiṣṭha, desired to sport with his wife on that mountain. And the king caused Vasishtha, to be proclaimed his regent in his capital and kingdom, in the woods and gardens. And bidding farewell unto the monarch, Vasiṣṭha left him and went away. Saṃvaraṇa, who sported on that mountain like a celestial, sported with his wife in the woods and the under-woods on that mountain for twelve full years. And, O best of the Bhāratas, the god of a thousand eyes poured no rain for twelve years on the capital and on the kingdom of that monarch. Then, O chastiser of enemies, when that season of drought broke out, the people of that kingdom, as also the trees and lower animals began to die fast. And during the continuance of that dreadful drought, not even a drop of dew fell from the skies and no corn grew. And the inhabitants in despair, and afflicted with the fear of hunger, left their homes and fled away in all directions. And the famished people of the capital and the country began to abandon their wives and children and grew reckless of one another. The people being afflicted with hunger, without a morsel of food and reduced to skeletons, the capital looked very much like the city of the king of the dead, full of only ghostly beings. On beholding the capital reduced to such a state, the illustrious and virtuous and best of Ṛṣis, Vasiṣṭha was resolved upon applying a remedy and brought back unto the city that tiger among kings, Saṃvaraṇa, along with his wife, after the latter had passed so long a period in solitude and seclusion. After the king had entered his capital, things became as before, for, when that tiger among kings came back to his own, the god of a thousand eyes, the slayer of Asuras, poured rain in abundance and caused corn to grow.”
As in the story of Devāpi, here the rightful king leaves kingship for the woods (although not for ascetic practice...) and so Indra stops rain for 12 years. Another passage, MBh I.89, mentions Saṃvaraṇa in relation to a great drought but also a war:
“While Saṃvaraṇa the son of Ṛkṣa, O king, was ruling the earth, there was a very great loss of people, so we have heard. The kingdom was shattered by manifold destructions in this way: struck by death for starvation, by want of rain and diseases, and the troops of the enemies attacked the Bhāratas. And shaking the Earth, so to say, with a fourfold army (i.e. made of chariots, elephants, knights and infantrymen), the Pañcāla marched against him (Saṃvaraṇa), and, having quickly conquered the Earth, he defeated him in battle with ten Akṣauhiṇis (troops of tenths of thousands of soldiers).”  
According to Pargiter, this Pañcāla king is none other than Sudās, defeating the alliance of the Ten Kings on the River Paruṣṇī (Ravi), where Saṃvaraṇa would be the king of the Pūrus. There is no mention here of sporting in the woods: the drought is associated with the rule of Saṃvaraṇa, and it is after the war that he takes shelter in the forests on the Sindhu, near the mountains. And there he is reached by Vasiṣṭha. A hint of the age of the son or descendant of Ṛkṣa is given also in RV VIII.68, where we find as patrons of the poet-priest a 'son of Ṛkṣa' and Indrota son or descendant of Atithigva, which is an epithet of Divodāsa, the famous ancestor of Sudās (with 5 generations of distance). Another hymn of the 8th book, RV VIII.74, mentions a Śrutarvan son or Ṛkṣa and also the river Paruṣṇī.

2) Viśvāmitra. MBh XII.139
"Towards the end of Tretā and the beginning of Dvāpara, a frightful drought occurred, extending over twelve years, in consequence of what the gods had ordained. At that time which was the end of Tretā and the commencement of Dvāpara, when the period came for many creatures superannuated by age to lay down their lives, the thousand-eyed deity of heaven poured no rain. The planet Bṛhaspati began to move in a retrograde course, and Soma abandoning his own orbit, receded towards the south. Not even could a dew-drop be seen, what need then be said of clouds gathering together? The rivers all shrank into narrow streamlets. Everywhere lakes and wells and springs disappeared and lost their beauty in consequence of that order of things which the gods brought about. Water having become scarce, the places set up by charity for its distribution became desolate. The Brahmanas abstained from sacrifices and recitation of the Vedas. They no longer uttered Vashats and performed other propitiatory rites. Agriculture and keep of cattle were given up. Markets and shops were abandoned. Stakes for tethering sacrificial animals disappeared. People no longer collected diverse kinds of articles for sacrifices. All festivals and amusements perished. Everywhere heaps of bones were visible and every place resounded with the shrill cries and yells of fierce creatures. The cities and towns of the earth became empty of inhabitants. Villages and hamlets were burnt down. Some afflicted by robbers, some by weapons, and some by bad kings, and in fear of one another, began to fly away. Temples and places of worship became desolate. They that were aged were forcibly turned out of their houses. Kine and goats and sheep and buffaloes fought (for food) and perished in large numbers. The Brahmanas began to die on all sides. Protection was at an end. Herbs and plants were dried up. The earth became shorn of all her beauty and exceedingly awful like the trees in a crematorium. In that period of terror, when righteousness was nowhere, O Yudhishthira, men in hunger lost their senses and began to eat one another. The very Ṛṣis, giving up their vows and abandoning their fires and deities, and deserting their retreats in woods, began to wander hither and thither (in search of food). The holy and great Ṛṣi Viśvāmitra, possessed of great intelligence, wandered homeless and afflicted with hunger."
The description reminds the one above from MBh I.63, but is more detailed, and is not focused on one town only, but on several towns and regions, we can suppose of North India. 

3) Sārasvata. The detail of wandering Ṛṣis reminds MBh IX.50, where the 12 years drought makes them forget the Vedas, but the young Sārasvata preserve them thanks to the Sarasvatī river: 
"a drought, O king, occurred that extended for twelve years. During that drought extending for twelve years, the great rishis, for the sake of sustenance, fled away, O monarch, on all sides. Beholding them scattered in all directions, the sage Sārasvata also set his heart on flight. The river Sarasvatī then said unto him, 'Thou needst not, O son, depart hence, for I will always supply thee with food even here by giving thee large fishes! Stay thou, therefore, even here!' Thus addressed (by the river), the sage continued to live there and offer oblations of food unto the Ṛṣis and the gods. He got also his daily food and thus continued to support both himself and the gods. After that twelve year's drought had passed away, the great Ṛṣis solicited one another for lectures on the Vedas. While wandering with famished stomachs, the Ṛṣis had lost the knowledge of the Vedas. There was, indeed, not one amongst them that could understand the scriptures. It chanced that someone amongst them encountered Sārasvata, that foremost of Ṛṣis, while the latter was reading the Vedas with concentrated attention.”
This story is a wonderful example of a crisis of cultural and even religious heritage due to a natural crisis, but the historical context is not clear. It is said that Sārasvata was son of the fabulous Ṛṣi Dadhīca. However, in Viṣṇu Purāṇa III.3, he is the Vedavyāsa (arranger of the Vedas) in the '9th' Dvāpara Yuga after Vasiṣṭha and before Tridhāman and Trivṛṣan: this last name corresponds to the name of the father of Tryaruṇa Traivṛṣṇa in RV V.27.1, and the Aikṣvāku king Tryaruṇa son of Tridhātu (a name taken from Pañcaviṃśa Brāhmaṇa 13.3.12, quite similar to Tridhāman, cp. Tridhanvan below) is a contemporary of Viśvāmitra according to Lloyd. We can add that Trayyāruṇa is the Vedavyāsa of the 15th Dvāpara Yuga. Apparently, there is some historical sequence in this list of Vedavyāsas, who are projected on a mythical sequence of different Dvāpara Yugas, but are rather to be placed in the same, the last, Dvāpara Yuga. 

4) Satyavrata Triśaṅku. According to Purāṇic genealogies, the son of Trayyāruṇa is Satyavrata Triśaṅku, the king who helped Viśvāmitra's wife during the famine. So says VP IV.3:
“Purukutsa had a son by Narmada named Trasadasyu, whose son was Sambhūta, whose son was Anaraṇya, who was slain, by Rāvaṇa in his triumphant progress through the nations. The son of Anaraṇya was Pṛṣadaśva; his son was Haryyaśva; his son was Sumanas; his son was Tridhanvan; his son was Trayyāruṇa; and his son was Satyavrata, who obtained the appellation of Triśaṅku, and was degraded to the condition of a Caṇḍāla, or outcast. During a twelve years' famine Triśaṅku provided the flesh of deer for the nourishment of the wife and children of Viśvāmitra, suspending it upon a spreading fig-tree on the borders of the Ganges, that he might not subject them to the indignity of receiving presents from an outcast. On this account Viśvāmitra, being highly pleased with him, elevated him in his living body to heaven.”

In this account, we have the mention of Rāvaṇa, the terrible king of Lanka, who in his raid in the subcontinent (or in Lanka itself) was vanquished by the Haihaya king Arjuna Kārtavīrya, then imprisoned and later released, and, as is well known, he was killed by Rāma Dāśarathi. As also Lloyd remarks, here we have a clear synchronism of Arjuna and Rāma, and so we have to place also Arjuna in the period of the transition between Tretā and Dvāpara; also his act of burning the earth with its settlements and the following invasion, together with Haihayas, of Western raiders (identified as Śakas, Yavanas, Kāmbojas, Pāradas and Pahlavas in Vāyu Puraṇa 26.121-128, invading the kingdom of Bāhu, 8 generations after Triśaṅku) can be connected to the same period. This harmonizes with the date (2000 BCE) given to the beginning of the so-called "Malwa Culture" (see here, p.227), which was also in Maheshwar (identified with the royal town of Arjuna, Māhiṣmatī), and with the presence of objects similar to Iranian ones in Malwa sites, like spouted pots (see here). Iranian or Central Asian affinities were found also in the Cemetery H culture starting from 1900 BCE. These western peoples can also be those mentioned in RV VII.18 like the Pakthas. So, this period of climatic crisis brought various invasions, especially from Central Asia, but also internal movements like that of the Haihayas from Malwa and of Rāvaṇa from the South. A drought is also connected with Lomapāda, the king of Aṅga in Eastern India, friend of Daśaratha, the father of Rāma (see MBh. III.310), but without mention of 12 years, that we find instead in the story of another figure contemporary of Rāma according to Lloyd, Māndhātṛ.

5) Māndhātṛ. MBh III.126:

“When there was a drought, which continued for twelve consecutive years, the mighty king caused rain to come down for the growth of crops, paying no heed to Indra, the wielder of the thunder-bolt, who remained staring (at him). The mighty ruler of the Gandhara land, born in the lunar dynasty of kings, who was terrible like a a roaring cloud, was slain by him, who wounded him sorely with his shafts. O king! he of cultured soul protected the four orders of people, and by him of mighty force the worlds were kept from harm, by virtue of his austere and righteous life. This is the spot where he, lustrous like the sun, sacrificed to the god. Look at it! here it is, in the midst of the field of the Kurus, situated in a tract, the holiest of all. O preceptor of earth! requested by thee, I have thus narrated to thee the great life of Mandhata, and also the way in which he was born, which was a birth of an extraordinary kind.”
It is significant that a Mandhātṛ is also cited in Ṛgveda (I.112, VIII.39, VIII.40), and he (called Māndhātṛ Yauvanāśva) is the poet of X.134 according to the Anukramaṇī, a hymn very similar to X.133 (sharing the meter mahāpaṅkti, phraseology and dedication to Indra), that is ascribed to Sudās and has also a refrain similar to RV VIII.39-40 and partially the same śakvarī and mahāpaṅkti meter. X.133 speaks of war, and the śakvarī meter is mentioned in the hymn of the battle of the Ten Kings, VII.33.4, as the meter of the cry that attracted Indra to fight on the side of Sudās (see the translation of Jamison and Brereton). So, we can suppose that they belong to the same age of Sudās, the king of the great battle, who is also a contemporary of Māndhātṛ in Lloyd's table. 

We will conclude with the story of Agastya: 

6) Agastya. MBh XIV.95:

"In olden days, O king, Agastya of great energy, devoted to the good of all creatures, entered into a Dīkṣā extending for twelve years. […] As Agastya, however, was engaged in that sacrifice of his, the thousand-eyed Indra, O best of the Bhāratas, ceased to pour rain (on the Earth). At the intervals, O king, of the sacrificial rites, this talk occurred among those Ṛṣis of cleansed souls about the high-souled Agastya, viz., 'This Agastya, engaged in sacrifice, is making gifts of food with heart purged of pride and vanity. The deity of the clouds, however, has ceased to pour rain. How, indeed, will food grow? This sacrifice of the Ṛṣi, ye Brahmanas, is great and extends for twelve years. The deity will not pour rain for these twelve years."
Agastya is known as brother of Vasiṣṭha, and is mentioned in MBh III.96, in connection with king Śrutarvan, Vadhryaśva (restored in the critical edition instead of Vradhnaśva) and Trasadasyu (III.96.12c-13a: agastyaś ca śrutarvā ca vadhryaśvaś ca mahīpatiḥ trasadasyuś ca). The famous Vadhryaśva is the father of Divodāsa Pañcāla, and Trasadasyu Paurukutsa can be the Aikṣvāku king already mentioned or, as Lloyd proposes, his descendant Tryaruṇa, called Trasadasyu in RV V.27.3.
Maybe it is due to this context that most of the hymns ascribed to Agastya in the Ṛgveda (I.165-191) end with the sentence: “May we know refreshment and a community having lively waters.” (vidyā́meṣáṃ vṛjánaṃ jīrádānum). And maybe significant is also his hymn to food and its juices (rasa), the so-called Annastuti (RV I.187), especially understandable in a period of famine and drought. 

So, we have finally to wonder if there is a possible scientific dating of this Great Drought. We have indeed some elements. In a paper by Berkelhammer et al. of 2012, "An Abrupt Shift in the Indian Monsoon 4000 Years Ago", we read: 
"Using a new high-resolution (~5 years/sample) speleothem stable isotope record from northeast India that spans the early and mid-Holocene, a number of abrupt changes in the oxygen isotopic composition of precipitation (δ 18 O p) are documented. The most dramatic of these events occurred ~4000 years ago when, over the course of approximately a decade, isotopic values abruptly rose above any seen during the early to mid-Holocene and remained at this anomalous state for almost two centuries. This event occurs nearly synchronously with climatic changes documented in a number of proxy records across North Africa, the Middle East, the Tibetan Plateau, southern Europe, and North America. We hypothesize that the excursion could represent a shift toward an earlier Indian Summer Monsoon withdrawal or a general decline in the total amount of monsoon precipitation. [...] the tight age constraints of the record show with a high degree of certainty that much of the documented deurbanization of the Indus Valley at 3.9 kyr B.P. occurred after multiple decades of a shift in the monsoon’s character but before the monsoon returned to its previous mid-Holocene state." "The most isotopically enriched values of the entire record occur between 4071 B.P. (±18 years) and 3888 B.P. (±22 years) during which the calcite remained enriched by ~0.8‰ relative to modern values (1.5‰ relative to the background values of the time) for a period of 183 years. The isotopic changes at this time manifested as a two-step process where values experienced a small steplike rise between ~4315 and 4303 years B.P. and experienced a second and more precipitous rise between ~4071 and 4049 years B.P. The abrupt shift occurred over approximately two decades, after which the values stabilized at this relatively enriched state for ~180 years before rapidly returning to previous background values at 3888 years B.P." "The monsoon over northeast India appears to have experienced an abrupt excursion at 4000 years B.P., the magnitude of which, in terms of both amplitude and length, exceeds any other event during either the most recent 600 years or throughout the early to mid-Holocene."

In this diagram from the paper (p.77) we see how around 4000 years BP we have a sudden fall of precipitation. In a supplementary list of dates the date of the sharp diminution of rain is 4055 BP, a sharp increase in 3941 BP, followed by another decrease, and finally a stable increase from 3899 BP. The lowest point is between 4049 and 4037 BP. Taken exactly, since BP is based on 1950 CE, this suggests a special drought in South Asia between 2099 and 2087 BCE. The error given is 30 years. Also a paper by Dixit of 2014 ("Abrupt weakening of the summer monsoon in northwest India ~4100 yr ago"), about the Kotla Dahar lake in Haryana, gives a similar date: "An abrupt 4‰ increase in δ18Oa occurred at ca. 4.1 ka, documenting a sharp reduction in Indian Summer Monsoon intensity."
So, it seems that the great drought of 12 years was in the 21st century BCE. We can also suppose that it is a stereoypical drought that refers to different episodes until 1900 BCE when precipitations came back (which can be alluded in the stories that stress the end of the drought). So, it can refer to several generations, but we have seen how the period involved appears to be mostly that between Triśaṅku and Rāma, that in my chronology is to be placed between the 21st and 20th century BCE. This drought is also placed in the same period (2200-2000/1900 BCE) where there was a strong aridity in the Near East and Central Asia, that likely caused migrations of people from the most arid regions, like the invaders mentioned in the Battle of the Ten Kings and in the Purāṇas. All this tends to confirm the historicity of several events described in the Mahābhārata and Purāṇas, and the chronology based on the date of Mahābhārata war in 1432 BCE. And denies that Indo-Aryans could have arrived during the 2nd millennium BCE, although some Iranic populations could have arrived into South Asia in that period.    

Update 30/1/2023

I have found today a 2018 article by Gayatri Kathayat et al., Evaluating the timing and structure of the 4.2 ka event in the Indian summer monsoon domain from an annually resolved speleothem record from Northeast India, that gives a similar picture but more precise and without the recovery of 'normal' monsoon. In comparison with the previous paper, it says:

"The 4.2 ka event in the KM-A record (Berkelhammer et al., 2012) manifests as a two-step change marked by an initial increase in the δ 18O values (∼ 0.6 ‰) between ∼ 4.31 and 4.30 ka, followed by another abrupt increase between ∼ 4.07 and 4.05 ka. The period between 4.05 and 3.87 ka in the KM-A profile is characterized by the most enriched δ 18O values over the entire record (∼ 1.5 ‰ higher than the background values before the event; Fig. 7), delineating ∼ 180 years of substantially weaker ISM. This multi-centennial period of enriched δ 18O values was terminated abruptly by a sharp return (< 20 years) to depleted δ 18O values, implying a resumption of stronger monsoon. The ML.1 and ML.2 δ 18O profiles during the contemporaneous period with the KM-A record, however, exhibit no step-like increase around ∼ 4.3 ka but instead an abrupt increase in the δ 18O values at ∼ 4.01 ka, which are superimposed over a gradually increasing trend over the entire length of the records. The timings and magnitude of this abrupt increase in the δ 18O values in both the ML.1 and ML.2 profiles are comparable to those observed in the KM-A profile (within the combined age uncertainties of both records; Fig. 7). A key difference between the KM-A, ML.1, and ML.2 δ 18O profiles, however, is the absence of a sharp decrease in the δ 18O values at ∼ 3.87 ka in our records, which marks the termination of the 4.2 ka event in the KM-A record.

The interval marking the onset of the 4.2 ka event in our record (∼ 4.255 ka) is indicated by a transition from pluvial (inferred by the lower δ 18O values) to variable ISM (dry–wet) conditions, with the latter superimposed by a few short-term (< decade) droughts (Fig. 8). Subsequently, the period between 4.07 and ∼ 4.01 ka is marked by persistently lower δ 18O values, implying stronger ISM (Fig. 8). The latter was terminated by a rapid increase in the δ 18O values (∼ 1.0 ‰, Fig. 5), suggesting an abrupt weakening of the ISM at ∼ 4.01 ka that occurred within a period of ∼ 10 years. Notably, as discussed above, the ML.1 and ML.2 δ 18O profiles show gradual increasing trends over the entire length of the record, which was punctuated by two multi-decadal weak monsoon events centered at ∼ 3.970 (∼ 20 years) and ∼ 3.915 ka (∼ 25 years), respectively (Fig. 8). These aspects of our ISM reconstruction differ from previous proxy records from the ISM domain, which typically portray the 4.2 ka event as a multi-century drought (e.g., Berkelhammer et al., 2012; Dixit et al., 2014). Our new data, however, demonstrate that prominent decadal to multi-decadal variability, together with the intermittent occurrence of multi-decadal periods of low rainfall, was the dominant mode of ISM variability during the period coeval with the 4.2 ka event."

So, here is confirmed a sudden and intense drought lasting approximately 10 years, around 2000 BCE. Here is the image showing a comparison with the previous analysis in the same cave, showing how the drought is placed later after a more humid period: the drastic change should have been catastrophic.


Sunday 13 January 2019

Mahābhārata and archaeology: the chariot of Sanauli and the position of Painted Grey Ware

Director SK Manjul from ASI showing the chariot from the excavation site at Sanauli

One of the great archaeological news of 2018 has been the discovery of the 'chariots' in Sanauli, UP, a village in the Ganga-Yamuna Doab where an archaeological site has been found, labelled as Late Harappan or Ochre Coloured Pottery/Copper Hoard. An interesting article from News18 says:
Spate of excavations in Uttar Pradesh’s Sanauli and Chandayana have thrown open possibilities of new cultures for the current archaeologists and they differ from Lal’s hypothesis. Lal correlated the ‘Painted Grey Ware Culture’ of archaeology to Mahabharata and bracketed it in the 1100 BC time frame. The new hypothesis now strikes the correlation between Ochre Coloured Pottery (OCP) with the Mahabharata Culture, which brackets it in 2000 BC. 
Based on the excavations conducted under him in Sanauli, where a chariot was found in June, SK Manjul, the director of Institute of Archaeology, under Archaeological Survey of India, delivered a lecture in July 2018 in Delhi. The title of the talk was “Mahabharata and Archaeology: PGW vis a vis OCP/Copper Hoard Culture.”

In the lecture, Manjul made a case for OCP as a culture relation to Mahabharata. The presentation was: “The hypothesis given by BB Lal and others on the basis of archaeological findings in the lower strata at Hastinapur and similar findings from other sites mentioned in the Mahabharata correlated Painted Grey Ware Culture with the Mahabharata Era.” Lal noticed Painted Grey Ware from important sites mentioned in Mahabharata and time bracket of war suggested around 1000-900 BCE. But recent excavations in Sanauli, Barnawa and Chandayana have a different story to tell - that of much earlier culture OCP. [...]  
As per the archaeologists, who are votaries of the New Hypothesis, PGW is marked with “rural settlements, pit dwellings and hut. There are weapons, arrow, head small, spear head [maybe arrow head and small spear head?], agriculture tools, bone points, metal iron. No chariots found. It shows limited and distinct pottery traditions. It has very less information on Vedic rituals and traditions.” 

On the other hand, the archaeologists say that the OCP is marked with “Advanced weapons and tools, antenna sword, harpoon, celts, dagger and shield, metal copper and advance chariot found. Advance pottery tradition including metal pots. Also, it has similarity with Vedic rituals.”
This connection was already proposed by S.P. Gupta and it was advocated also in a 2017 article by Vinay Kumar Gupta and B.R. Mani about Painted Grey Ware. I have also supported it in my article about the chronology of Mahābhārata and Ṛgveda, after having determined that the great battle should be placed in the year 1432 BC, a date based on the astronomical details from the poem and on the recurrent Purāṇic statement about the 1015 or 1050 years between the birth of Parikṣit (contemporary with the battle) and the coronation of Mahāpadma Nanda that can be placed in 417 or 382 BC (see the article for more details). About the archaeological connection, I wrote there: 
If we look at archaeology in order to find a corroboration of this date, we cannot reasonably pretend to find traces of the battle, but we do have some interesting elements for comparison. In the late 15th century B.C. in the area of Kurukṣetra we still have Late Harappan settlements, and in one of them, Bhagwanpura, particularly close to the supposed site of the battle, we find Late Harappan pottery together with Painted Grey Ware, dating here from 1400 B.C.[i] It is well known that according to B.B. Lal this ware is related to the Mahābhārata period because it is found in many localities mentioned in the poem, like Indraprastha (the capital of the Pāṇḍavas, identified with Purana Qila in Delhi), Hastināpura (the capital of the Kauravas), Ahicchatrā (the capital of North Pañcāla) and Kauśāmbī (the capital of South Pañcāla). But we have another ware that is present in the area of the Kurus (Upper Doab) and in the same sites of Hastināpura, Ahicchatrā and Kauśāmbī in the II millennium B.C., namely the so-called Ochre Coloured Pottery, as already observed by S.P. Gupta.[ii] This is the pottery of the first level of the aforementioned sites, it is generally dated 2000-1500 B.C., but the thermoluminescence tests on sherds from Atranjikhera, Lal Qila, Jhinjhana and Nasirpur have given dates between 2650 and 1180 B.C.[iii] About Hastināpura and Kauśāmbī, there is the important tradition that the fifth successor of Parikṣit, Nicakṣu, abandoned the first city, because it was carried away by the Ganges, and made the second one his capital.[iv] B.B. Lal has claimed that this is confirmed by the PGW levels of Hastināpura, where there are traces of a partial flood, and by the fact that we can find a similar PGW culture in Kauśāmbī. But we can observe that also the first, OCP level, of Hastināpura was abandoned, and that also in Kauśāmbī there are OCP levels which have been only hypothetically dated by Sharma in 1960.

[i] Kenoyer (2006, 43).
[ii] Gupta and Ramachandran (1976, 47-9).
[iii] Ghosh (1989, vol. I, 174-5).
[iv] Pargiter (1922, 285); Singh (2004, 143).  

I also noticed that PGW is absent from Gujarat, although places like Dvārakā and Prabhāsa in Kathiawar have a central role in the poem in connection with Kṛṣṇa and the Yādavas. And, if Kṛṣṇa's Dvārakā is to be identified with the submerged city found by S.R. Rao near Bet Dwarka, and, as he said, “it is possible to postulate on structural, ceramic and inscriptional evidence that Dwarka was built in the 15th century B.C. when the sea level was lower than at present and was submerged within a hundred years”, we have another confirmation of this chronology, because according to the poem (see here) the city was submerged 36 years after the battle, that is, in 1396 BC, and at the same time the remaining Yādavas abandoned Kathiawar towards Indraprastha after the massacre of their clans called Vṛsṇi, Andhaka and Bhoja. According to Kenoyer, most Late Harappan sites in Gujarat have a break after 1400 BC. Also Mukhtar Ahmed in a book on the end of the Harappan Civilization observes that the Lustrous Red Ware (that was found also in Bet Dwarka) continues until around 1400 BC, and afterwards there is a break in the archaeological record at most sites until around 600 BC. The coincidence is really impressive, and since it appears to confirm so precisely the account of the Mahābhārata we wonder if there are material traces of the migration. Mukhtar Ahmed says that Lustrous Red Ware was found also in Navdatoli III and Ahar IC, in the Ahar-Banas culture. Pargiter (p.279) wrote that the river Banas or Parṇāśa is connected with the Yādava king Devāvṛdha, brother of Andhaka and Vṛṣṇi, and his descendants reigned at Mārttikāvata, "which was apparently in the Śālva country around Mt. Abu", that is, in South Rajasthan where the Ahar-Banas culture was. And Arjuna established one of the emigrating Yādava princes, the son of Kṛtavarman, in Mārttikāvata. 

There was another kind of pottery common between Gujarat and this region, namely, the Black and Red Ware (BRW), that had a very long tradition in the Ahar-Banas culture, especially in Balathal, from the 4th millennium BC. The same kind of ware is dated 1450-1200 BC in the Western Ganges plain. This age of start is remarkably close, although there is also the theory that it came from the BRW tradition of the Eastern Ganges valley. As the Japanese archaeologist Uesugi has written (see here): "During the second millennium BCE, the black ware industry was present in the Ganga Valley and southern Rajasthan and made its appearance in the western Ganga Valley and northern Rajasthan, to which the black ware industry was introduced either from the eastern Ganga Valley or from southern Rajasthan, as this region was widely occupied by the Bara‐OCP complex during the early second millennium BCE. There is no conclusive evidence for determining from which region the black ware industry [was] introduced to the western half of the Ganga Valley, but the morphological features of BRW and BSW [Black Slipped Ware] from this region suggest their introduction from the eastern Ganga Valley rather than from southern Rajasthan." However, later on he says: "Noteworthy is that BRW and BSW in the western Ganga Valley include shallow bowls/dishes in their formal assemblage (Figure 9: 22, 23). The origin of this form cannot be revealed with the currently available evidence in the Ganga Valley, but as it was absent in the assemblage of the eastern Ganga Valley during the early second millennium BCE (Purushottam Singh 1994, 1996), it can be presumed that this form emerged during the late second millennium BCE in the Ganga Valley." But shallow bowls were present in the Ahar-Banas BRW (see here and here).

So, if BRW came to Western UP from Rajasthan, we would have a sign of the migration of the Southern Yādavas, taking their seat in the area of Indraprastha in the Doab, where Arjuna brought "the bulk of the people", as Pargiter says and installed Vajra as king. On the other hand, in the Viṣṇu Purāṇa Arjuna establishes Vajra as king in Mathurā, which is much more logical, since Mathurā was already a Yādava capital before the invasion by the king of Magadha Jarāsandha that compelled Kṛṣṇa to go to Dvārakā.

Uesugi continues: "In the western Ganga Valley and northern Rajasthan, the black ware industry is identifiable as having an independent phase between the Bara‐OCP phase and the PGW phase and as continuing to the following PGW‐dominant phase (Gaur 1983). Even in the Ghaggar Valley, BRW and BSW are known to be associated with PGW, although there is no independent phase of BRW and BSW in this region. These pieces of evidence imply that the black ware industry and PGW had some relations."

So, the Painted Grey Ware in the Doab comes after the BRW, and in the Ghaggar valley they come together. This means that the PGW culture is connected with that local tradition, and not with an alleged western migration. What Uesugi also remarks is that the PGW pottery is radically different from the previous Bara-OCP pottery of Late Harappan affinity for the forms and the painting motifs. He also doubts that the overlap of PGW and Late Harappan in Bhagwanpura is real, because it concerns only potsherds and not full vessels in burials. However, even if we accept the overlap, we have to do with two different traditions. Was the PGW culture the result of the arrival of the southern Yādavas? According to Uesugi, the area of origin of PGW can be the Ghaggar valley: "The region of origin of PGW has not been specified, but the dense distribution of PGW sites in the Ghaggar Valley and the cultural sequences in different parts of North India suggest that PGW developed in the Ghaggar Valley. However, it is important to repeat that PGW did not have direct relations with the Bara‐style pottery which was widespread in the preceding period but had connections with the black ware industry in the Ganga Valley. Therefore, the origin of PGW must be searched in its relations with the black ware industry in the east, that is in the connection between the Ghaggar Valley and the Ganga Valley during the second millennium BCE. This is a hypothesis to be tested against more evidence in future studies." 
I find quite difficult to accept that the PGW culture could develop in the Ghaggar valley where the local tradition was completely different. It seems more likely that it developed where there was already the similar BRW-BSW. Another region with high frequency of PGW sites is the region of Mathurā or Braj region, as noticed especially by the already cited article by Kumar Gupta and Mani: "The Braj region for sure presents the most dense settlement pattern in the entire sub‐continent and the area west of Yamuna has to be the core area of the PGW where the minimum area required for the presence of a PGW settlement comes out to be just about 16 sq. km. [...] Now, we have quite solid grounds to speak about the authors of the PGW culture. The PGW using people were indigenous to the area where their settlements are found in high density, i.e., the Braj region. They did not come from outside. Their pottery technology seems to be an advancement over the pre‐existing Black Slipped Ware and plain grey ware technology, so they were in continuity of the earlier folks who used Black Slipped Ware with grey core and plain grey ware."

Painted Grey Ware

If the Braj or Mathurā region was the place of origin of the PGW, it would be of Yādava origin. Anyway, it is clear that the PGW culture spread in the Kuru-Pañcāla kingdom that was central in the late Vedic period of Brāhmaṇas, a more peaceful culture compared to the Ṛgveda: in the same way, PGW culture seems less warlike than OCP/Copper Hoards, as the same article remarks: "The Mahabharata was a period of continuous warfare and the limited weaponry evidence from PGW horizons does not favour it whereas Copper hoards are associated with OCP on a secure ground and the Copper hoard implements might be indicative of an atmosphere of warfare."
In fact, the new discoveries from Sanauli/Sinauli show us a warrior culture with swords and chariots, but this culture is not only typical of the Mahābhārata: also the Ṛgveda mentions battles, warriors, chariots, and sometimes swords and spears (asi and ṛṣṭi). If the age of the Sanauli chariots is 2000-1800 BC (or also 2100 BC as I have heard from another source) it would correspond to the age of the early Ṛgveda according to my chronology. Finally, we would have the concrete representation of the rathas of those ancient hymns.
Update April 2020: While preparing an article from this post, I have deepened some topics, so that I have realized that Uesugi must be correct: BRW came from the Ganges valley to Western UP and this coincides with the great rising of the power of the Magadha king Jarāsandha around 1450 BC, which compelled also other populations to flee towards the west, as clearly said in the Mahābhārata (II.13, see here):
At present, however, O monarch, king Jarasandha, overcoming that prosperity enjoyed by their whole order, and overpowering them by his energy hath set himself over the heads of all these kings. And Jarasandha, enjoying the sovereignty over the middle portion of the earth (Mathura), resolved to create a disunion amongst ourselves. O monarch, the king who is the lord paramount of all kings, and in whom alone the dominion of the universe is centered, properly deserves to be called an emperor.
And, O monarch, king Sisupala endued with great energy, hath placed himself under his protection and hath become the generalissimo of his forces. And, O great king, the mighty Vaka, the king of the Karushas, capable of fighting by putting forth his powers of illusion, waiteth, upon Jarasandha, as his disciple. There are two others, Hansa and Dimvaka, of great energy and great soul, who have sought the shelter of the mighty Jarasandha. There are others also viz., Dantavakra, Karusha, Karava, Meghavahana, that wait upon Jarasandha. [...] And, O king of kings, Bhishmaka, the mighty king of the Bhojas--the friend of Indra--the slayer of hostile heroes--who governs a fourth part of the world, who by his learning conquered the Pandyas and the Kratha-Kausikas, whose brother the brave Akriti was like Rama, the son of Jamdagni, hath become a servitor to the king of Magadha. [...] And, O exalted one, the eighteen tribes of the Bhojas, from fear of Jarasandha, have all fled towards the west; so also have the Surasenas, the Bhadrakas, the Vodhas, the Salwas, the Patachchavas, the Susthalas, the Mukuttas, and the Kulindas, along with the Kuntis. And the king of the Salwayana tribe with their brethren and followers; and the southern Panchalas and the eastern Kosalas have all fled to the country of the Kuntis. So also the Matsyas and the Sannyastapadas, overcome with fear, leaving their dominions in the north, have fled into the southern country. And so all the Panchalas, alarmed at the power of Jarasandha, have left their own kingdom and fled in all directions. Some time before, the foolish Kansa, having persecuted the Yadavas, married two of the daughters of Jarasandha. They are called Asti and Prapti and are the sister of Sahadeva. Strengthened by such an alliance, the fool persecuting his relatives gained an ascendency over them all.

The movement from Magadha and generally from the east to the west corresponds to the movement described by Uesugi about the diffusion of BRW. The Japanese archaeologist also maintains that PGW formed later, with eastern influence, in the 'Upper Ganges' region, that includes the Ghaggar valley and corresponds to the land of the Kurus, which is central in the late Vedic period. On the other hand, the Indian archaeologist Vijay Kumar (see here) has recognized also patterns from the southern Ahar tradition in PGW pottery: "The PGW has been tentatively assigned the time bracket of 1200 to 600 B.C. The diagnostic pottery of this phase is PGW. The shapes are similar to those found in BRW phase. The new painting designs are introduced during this phase. Roughly linear parallel lines on grey pottery are inspired from corded ware tradition of the east but the curved and wavy lines are inspired from similar paintings on Aharian black and red ware." 
So, it is possible that also a Yādava influence from Rajasthan was present in the PGW tradition, that started after the great migrations following the Mahābhārata war. The long Harappan-OCP tradition was now replaced by a rising Gangetic culture, as is also stated in MBh. III.85: “It hath been said that in the Satyayuga all the tirthas were sacred; in the Treta, Pushkara alone was such; in Dwapara, Kurukshetra; and in the Kali-yuga, the Ganga alone is sacred.”