L1a2-L1307, L1a2-M357 or in older publications L1c is one of the oldest, most widespread, and diverse haplogroups in Afghanistan and South Central Asia. It can also frequently be found in the Caucasus among people like Chechens, in East Anatolia and in the northwestern part of the Indian Subcontinent among people like Jatts, Rors, Khatris, and Sindhis. But it is much less common in India east of Punjab. In lower frequencies, it is generally found from Syria to West China (Xinjiang). The historical importance of this line in the history of Afghanistan can not be understated. It had formed around 17.000 years ago, but so far all modern and ancient L-L1307 carriers fall under L1307 branches that split around 8000 years ago (TMRCA) in the late Neolithic. It has a frequency of around 10% today among Afghans with some Pashtun or Baloch tribes reaching more than 50% so being dominated by that clade. This is especially true for Durrani Pashtuns and their subtribes, who seem to be dominated by it. Durrani Pashtuns, together with some Baloch and Brahuis, fall under the L-Y61970 branch, which just formed around 2000 years ago. This is most likely the result of a massive founder effect that can be dated to the historic or medieval era. It is also very likely that the royal Barakzai dynasty falls under this haplogroup.
The ancient origins of L-L1307 are not fully clear yet and the exact homeland of L-L107 is disputed. It was not found in early Neolithic groups from the Iranian Plateau, but became in the late Neolithic common in the Iranian Plateau and Afghanistan. The oldest L1a was found in Chalcolithic Armenia and Turkmenistan. In Areni (Vayots-Dzor Province), Armenia L1a1 a branch parallel to L1a2 was found (3800-4200 B.C). L1a1 split already in the Mesolithic around (17.000 years ago) from L1a2, but still considering the general lack of L1a2 in DNA older than the late Neolithic this is a very interesting data point. The 3 Armenian samples with L1a1 were of mainly East Anatolian ancestry rich in Neolithic Anatolian ancestry, but they also had significant Neolithic Steppe, Caucasus, and to a lesser extent Neolithic Iranian ancestry or affinity. So it is not fully clear with which ancient ancestry component L1a was initially associated but the most likely candidate seems to be Neolithic or Mesolithic East Anatolian and North Mesopotamian ancestry (Laziridis 2016).
Another data point is a L1a sample from from late Neolithic Monjukli-Depe in Turkmenistan (dated around 4600 B.C). We have no autosomal for this sample yet because the genetic data was not released yet (Allentoft 2022). But it is likely that this ancient sample clustered with already published samples from Tepe Anau and Namazga dated bit later. Other possible hints to the ancient origin of L1a2 are many basal L1a2 clades in the Near East and Caucasus. These L1a2-L1037 clades don’t look like recent arrivals from Central or South Asia because they split from most South-Central Asian clades already in the Neolithic. L-L1307 was for example found in Iron Age (around 900 B.C) Hasanlu in modern-day northwestern Iran (Lazaridis et al. 2022).
I4233 : (Hasanlu, Iran)
Interestingly this ancient L-L1307 sample was closest to modern day Mesopotamian and Caucasian Jews, because it was rich in East Anatolian, Neolithic Iranian and also Neolithic Levantine ancestry.
So both ancient and modern L1a(2) point to L1a2 arriving in Afghanistan and South Central Asia during the late Neolithic from a region between north Mesopotamia and the South Caucasus. This population was likely rich in Iran Neolithic admixture but genetically significantly more western than earlier Pre-Pottery populations of the Iranian Plateau (Ganj Dareh, Wezmeh cave, Hotu cave, Abdul Tepe Hosein). The exact timing is not clear yet but for most clades was likely between 4000-6000 B.C with some L1a2 arriving from much later immigration waves from West Asia.
Late Neolithic pottery in North Mesopotamia and West Iran:
So it is not surprising that almost every group in Afghanistan and South Central Asia has old L1c subclades splitting from other L1c clades of nearby ethnic groups in the late Neolithic and earlier. This is different from most R1a and Q1b clades, which only started to massively expand during the middle and late Bronze Age, and are spread over huge territories with relatively young TMRCAs around 1500-3000 BC. On the other hand, most of these L-L1307-rich ethnic groups show ethnic-specific founder effects for their specific clades during the Iron Age and historic era. So the diversity of L-L1307 clades within an ethnic group or tribe can often be quite low. Summarized, it looks like L-L1307 had massive growth and diversification during the late Neolithic, then decreased in frequency and diversity during the late Bronze, and later again expanded massively in some ethnic groups and tribes during the Iron Age and historic era.
The oldest ancient L-L1307 dated to around 3000 BC was found at the Bronze Age Shahr-i-Sokhta site in modern-day Sistan, Iran. The sample with the ID I11462 was not just positive for L1307, but also for L-Z20267. On the other hand, it was negative for L-Y12419 under which the big Durrani Pashtun and Baloch clade falls, so I11462 is likely not a direct ancestor of most (South) Afghan L-L1307 today. Still the location and historical importance of Shahr-i-Sokhta makes it likely that at least some of the Afghan L-L1307 derive from this and similar sites in the region. It is also worth mentioning that this sample and most other samples of the archaeological site had autosomal DNA profiles rich in Iran Neolithic but also more western Iran Chalcolithic ancestry. They likely had Neolithic Iranian ancestry from different immigration waves with one later immigration wave bringing L1a2 to the region, but the exact timing of the arrival remains unclear. This ancient L-L1307 sample in terms of modern affinity would be closest to modern-day Balochs and Gedrosians, but still quite distinct from these modern populations because it lacked Bronze Age Steppe ancestry and had lower South Asian admixture. But some of the other samples in Shahr-i-Sokhta had a lot of South Asian ancestry and even clustered with modern-day Central Indians. Such samples likely represent migrants from the Indian Subcontinent rather than local individuals.
I11462: (Shahr-i-Sokhta, Iran)
406, IRV Grave 309 (I11462): Context date of 3000-2900 BCE. Genetically male. Single burial (Period I, phases 10 or 9) in a catacomb grave (vertical shaft at the southeast, lateral chamber at the northwest, closed with a mudbrick wall). The deceased (Individual A, age at death not determined) was placed with flexed lower limbs on the right side. A broken pottery bowl was placed at the bottom of the shaft in front of the wall. A pottery beaker, a banded travertine bowl, and two beads (in turquoise and red limestone, respectively) were
found in the chamber near the cranium.
Shahr-i-Sokhta is an archaelogical site of the Helmand culture flourishing between 3200-2350 BC. During this period, it was one of the biggest cities in the world and now the archaeological site contains between 20.000-40.000 graves. It is located close to the eastern edge known as the Lut Desert, one of the hottest places on Earth today. But the climate was far more hospitable for agriculture in ancient times when the Hamun Lake and marshlands (near which Shahr-i-Sokhta was located) were much greater in size.
Shahr-I-Sokhta artifacts:
Considering its big size and richness, it is not surprising that Shahr-i-Sokhta showed archaeological links to sites in Elam/Iran, Turkmenistan and even the Indus region. Like mentioned before, there has also been DNA evidence of people from the Indus Valley migrating to Shahr-i-Sokhta. But Shahr-i-Sokhta was not the only sizeable site of the Helmand culture. Other famous sites include Mundigak in the Kandahar province for example. But Shahr-i-Sokhta has so far been the only archaeological Bronze Age site from the Balochistan and South Afghanistan region that has been studied for archaeogenetics. It is possible that the ancestors of much of the South Afghan and Balochi L-L1307 already lived near other archaeologically excavated sites in the region (shown in the map below).
Two other old L-L1307 samples were found in archaeological sites related to the BMAC (Bactria-Margiana Archaeological Complex) culture. Both samples were dated to around 1900-1700 BC and postdated already the heydays of the BMAC culture, but culturally and genetically they were closely related to earlier BMAC cities like Gonur.
I5604: (Bustan, Usbekistan)
UZ-BST-011, Site 4, Grave 26, 57-030 (I5604): Date of 1880-1697 calBCE (3465±20 BP,PSUAMS-2774). Genetically male. This individual is genetically detected as a 2nd or 3rd degree relative of I11519
I4285 : (Sappali Tepe, Uzbekistan)
UZ-ST-003, Sappali Tepe (ST) 1971, 38, Grave 00-95 (I4285): Date of 1873-1661 calBCE (3430±25 BP, PSUAMS-2536). Genetically male.
Both samples seem to belong to basal, and probably rare today, clades. But like in the case of Shahr-i-Sokhta, the (Post)-BMAC cultures to which the samples belonged, likely played a significant role in the formation of the Afghan gene pool.
The intensification of agriculture and development of irrigation during the Chalcolithic and early Bronze Age caused massive growth and centralization of settlements, culminating with the blossoming of the Bactrian-Margian Archaeological Complex (BMAC), also called the Oxus Civilization (Lyonnet and Dubova, 2020). BMAC shows the first structured proto-urban cities of the area gathering thousands of individuals, and a deep social structuring in Central Asia (Muradov, 2021). It suddenly appeared in Central Asia and then quite quickly disappeared for regions not fully understood. Genetic data obtained from several archaeological sites of BMAC and Post-BMAC show some genetic continuity between the late Neolithic and BMAC but with immigrants from Iran, South and Central Asia settling in the BMAC culture. Most of the ancestry of BMAC-related people was like that of Shar-i-Sokhta, derived from Neolithic groups of the Iranian plateau (50-80%). But a significant part of that ancestry had more recent origin in West Asia ( Iran Chalcolithic ancestry) and some samples from the Post-BMAC site of Dzharkutan even were recent immigrants from the South Caucasus.
BMAC artifacts:
Another ancient L-L1307 hotspot was the Swat Valley (Swat and Dir districts) in modern-day Pakistan, which was also studied for archaeogenetics. Here, L-L1307 was found in 14 Iron Age samples. But like in the case of the L-L1307 samples from BMAC and Shahr-i-Sokhta, most of these L-L1307 were either low coverage or fall under basic clades splitting from most modern and Afghan L-L1307 already in the Neolithic. Still I8220 from the Aligrama site dated to 756-430 BC was positive for Y12419 under which the big Abdali Pashtun L-L1307 also falls. The TMRCA of this haplogroup is still around at 4300 BC, so again probably not a direct ancestor, but still the closest ancient Y-DNA sample for Durrani L-Y61970. However much of Afghan L-L1307, especially in eastern Afghanistan, was not deeper analyzed, so closer matches may be found in the future.
Iron Age and Historic era Swat were also rich in Neolithic Iranian ancestry (on average around 50%) but had obviously deeper connections to the Indian Subcontinent than BMAC. Their West Asian ancestry was not that western shifted like that of BMAC and Post-BMAC and rather similar to that of earlier Eneolithic groups and early Bronze Age groups from Turkmenistan and Shar-I-Sokhta. Swat was an ancient centre of urbanism, Sanskrit, and Buddhism. But based on genomic and historical evidence, we can assume that a large non-Indo-Iranian population coexisted with Indo-Aryans, hence the low Steppe admixture and low frequency of typical Steppe lines R1a-Z93. People from ancient Swat migrated in significant numbers into Afghanistan and Central Asia during the Buddhist and Kushan period, but would in the medieval period be replaced by Pashtun settlers from the southwest.
Interestingly ancient DNA studies have found multiple South Central Asian-related or admixed samples from Iron Age or medieval Central Asia and Xinjiang with L-L1307, often under very basal clades. One genetic outlier from Kazakhstan, who likely was a recent immigrant from the south, even clustered with modern day South Pashtuns (KNT005,Kazakhstan – Otrar-Karatau culture dated to around 300-400 AD).
It seems that until the historic era, Y-DNA L1a (mostly L-L1307) was one of the most frequent haplogroups among agriculturalists from the Pamir Mountains to the coastal region of Makran. Pastoralists on the other hand probably tended to carry more R1a and Q besides a lot of L and especially L-L1307 as well.
Sources
- Lazaridis, Iosif, et al. “The genetic structure of the world’s first farmers.” bioRxiv (2016): 059311.
- Allentoft, Morten E., et al. “Population genomics of stone age Eurasia.” bioRxiv (2022): 2022-05.
- Lazaridis, Iosif, et al. “The genetic history of the Southern Arc: A bridge between West Asia and Europe.” Science 377.6609 (2022): eabm4247.
afghandna.wordpress.com 
Leave a comment