Afghan DNA discussion

[Negah]

You just kind of said the same thing as before, which I replied to before. My comment is almost entirely about the south asian in pashtuns.

Pashtuns can't be compared with hindukush kalash, the study samples Naramsinhan has access to. Kalash are inbred, pashtuns aren't.

There's not a study for every single thing, so I can't provide a study making this kind of specific comments about example pashtun haplogroups. If there were, I would had posted them.


https://x.com/Afghan_DNA/status/1789785032179012078

https://x.com/Afghan_DNA/status/1946508818470547604

https://x.com/Afghan_DNA/status/1944146672407523671


These are private and study DNA samples of pashtun haplogroups. If you look at their clades on yfull, the Z2123, L1c, R2 and H subclades, you can see they're rather closest to brahuis, jatts and split from them latest. Not anyone in hindukush mountains. That's not for no reason. Those subclades especially don't exist in kalash, who are very inbred. Pashtuns shows clear evidence of mixing with Suleiman mountains indics. This can be further supported based on linguistics, with the significant lahnda punjabi dialect influence on pashto language. Lahnda is spoken in parts of Suleiman mountains too. All pashto dialects shows this linguistical influence.



Arrian in his Anabasis refers to the Suleiman mountain locals, using sources from greeks in 300-200 BCE, as "indian mountaineers"

https://archive.org/details/cu319240...up?q=Arachosia

This is considering the fact that Arachosia included the Suleiman mountains.



Hsuen Tsang in 630 CE refers to the local languages spoken in much of Suleiman mountains to been "indian languages", except in the areas where pashtun lived at the time (they only had a small part of the mountains)


Even in 1030 CE, Biruni refered the locals of northern Balochistan as "savage hindus", despite mentioning pashtuns living north of those indics


https://archive.org/details/alberuni...e/198/mode/2up


Eventually pashtuns took over half of the mountains, even Waziristan. This isn't based on impressions or mythology, this is based on historical attestations. The locals in the mountains, prior pashtuns arriving, were simply indics. And clearly akin to lahnda speakers, such as saraikis, who still lives in eastern parts of the mountains. Those are just regular punjabis.



https://archive.org/details/Shahpur/page/412/mode/2up (inscription about pashtuns is from 270 CE)

https://archive.org/details/62617658.../1up?q=afghans (letters are from 480 CE)



As we can see here, pashtuns, originally called afghans, were living right next to white huns, who lived right north of Hindukush and Hazajarat. The sassanid inscription simply makes it clear that greek "Abgan" is actually awghan, comparing it to Middle Persian inscription. There's just no w or gh letter in greek script, so we can luckily compare it to non-greek inscriptions. Awghan is how pashtuns natively says afghan



https://astrofoxx.wordpress.com/wp-c...-of-2_1981.pdf (540 CE)

Varahamihira, a brahmin of magian descent (likely from north Afghanistan originally), refered white huns and afghans (pashtuns) twice in his book, showing them geographically connected to each other. This helps further to strengthen the fact that pashtuns were in North Afghanistan before moving to Suleiman mountains, and very much the same as the "abgans" refered in bactrian letters. Pashtuns were most rather likely in Hazarajat, as pashtuns are clearly yaz iranic descended since their R1a pashtun specific clade split from kurds in 1200-1100 BCE. Hazarajat lies closer to the yaz split area.


Attachment 144856


Attachment 144857


On top of that, Varahamihira separates pashtuns from cavemen. That's refering to borderlands, as shown in the page, since we had indics living in the mountains. The mountains in question would be the Suleiman mountains, which has for 1000s of years been considered a frontier between India and Khorosan, unlike the hindukush mountains. That means pashtuns according to Varahamihira, werent in Suleiman mountains yet. And unlike with greek and Middle Persian, the indic script Varahamihira used had both a b and a v letter. He used the v letter, and wrote afghans in plural as "Avagana". There's no gh letter in indic scripts though at the time. Again, an attestation, not myth.





It's only in 630 CE we see pashtuns in Suleiman mountains, seemingly only in Khost, parts of Paktya and Logar, they were in. Hsuen Tsang passed much of the mountains, and only came across the nation of "O-Po-kin", rendered to "Avakan" (closest rendition of Awghan)


https://dn790003.ca.archive.org/0/it....umich.edu.pdf



Attachment 144858


But prior Hsuen Tsang, we never hear of anything about pashtuns in the mountains historically.



Even linguistics support this from north to south migration. Pashto shares several cognates with other iranic languages, which has turned the D letters into L, which is also called lambdacism. This is a trait in some hindukush and pamiri languages, along with bactrian, which is actually a bactrian feature. Such feature only took place within hindukush and seemingly hazarajat:



https://www.academia.edu/49862012/La...an_t_in_Pashto


Attachment 144859

Attachment 144860



And even then also, we see pashtuns in the bactrian letters and sassanid inscription with iranic names, which would be out of character for non-muslim hindu indians. Even Hudud al alam also separates non-muslim pashtuns from hindus, buddhists (idolators) and muslims. This proves pashtuns weren't hindus before islam, but rather iranic pagans/zoroastrians. The actual locals of Suleiman mountains, as even Babur refered the remaining ones, would been hindus. So pashtuns came from outside Suleiman mountains, settling amongst the local indic hindus


https://archive.org/details/in.ernet...e/2up?q=afghan

Attachment 144861



The haplogroups, along with the history and linguistics, clearly shows pashtuns came from outside Suleiman mountains, most likely northwestern hazarajat (lies closer to yaz descended populations, as hindukush locals dont have yaz clades), possibly brought or forced out of the mountains into Nangarhar, Kabul by turk shahis in 550s, and then finally settled into the Suleiman mountains for 80 years prior Hsuen Tsang.

They would 100% not have such Suleiman local indic subclades.


So no, it isn't crazy to say pashtuns mixed with indics in Suleiman mountains and increased their IVC. Even tajikistanis in another study shows per DATES, going through a late event of significant south asian admixture in 1000 CE, most likely indian slaves brought in the towns


https://www.nature.com/articles/s41598-021-04144-4


Attachment 144862


So what's crazy about pashtuns mixing with local indians in the mountains? Their haplogroup and language shows mixing?

Thanks. I will read this carefully and respond.

[thisismyaccount]

@thisismyaccount thanks for responding, but I need to be very clear with you.

Your entire argument, it seems to me, is based on the idea that Pashtuns picked up 30–50% “Indian” ancestry recently (500–1000 CE). IMO that claim collapses the moment you look at actual genome-wide data. I ask you to do it objectively.

Again, I’m not asking you to agree with me. I’m asking you to read the sources and then bring sources of your own. Otherwise we’re just trading personal beliefs. Genetics does not work like that.

1) BMAC already had Indus-related ancestry 2500–2000 BCE. I ask you to please read my previous post, item 6c)

This is not my opinion. This is ancient DNA.

Narasimhan et al. 2019 (Science):
https://www.science.org/doi/10.1126/science.aat7487

“Three BMAC individuals dated 2500–2000 BCE from Gonur show elevated proportions of AHG/Indus Periphery ancestry.”

So the IVC-related ancestry that Pashtuns have was already flowing north into the Iranian Plateau and Central Asia thousands of years before the medieval period.
This is the prehistoric source of the South Asian-related ancestry in Pashtuns and certainly not medieval Punjabis.

2) AASI in the Hindu Kush is ancient, not recent

You completely misinterpreted Narasimhan. Please read him and if you don't agree, don't state your opinion. Your opinion has value if it is backed by reputable sources.

He literally says the opposite of your point:

https://www.science.org/doi/10.1126/science.aat7487

“The AASI ancestry in Hindu Kush groups reflects deep, pre-Bronze Age layers not attributable to recent South Asian admixture.”

This is the key line.
If your model were correct, Narasimhan’s qpAdm runs would show a recent Indo-Gangetic signal. They do not. They show ancient ANE+AASI highland ancestry. You do realize this is the stuff that predates Indo-Iranians themselves. Again, check the timelines that I provided.

So again: you need a source that models Pashtuns with recent Punjabi/Sindhi input. Without that, this remains a belief.

3) Y-DNA clades do NOT measure % ancestry

You keep using haplogroups (L, H, certain R1a branches) to argue 30–50% “Indic.”
That isn’t how population genetics works.

Hellenthal et al. 2014 (Science):

“Y-chromosome admixture proportions are typically very different from autosomal ancestry proportions.”

Y-DNA tracks one male line, not full ancestry.
Autosomal qpAdm is what counts, and those models reject recent Indo-Gangetic ancestry in Pashtuns.

4) Your interpretation of AASI is historically incorrect

AASI is not “Indian.”
It is 30,000–10,000 years old, predating Indo-Aryans, Indo-Iranians, and even early IVC. It existed across the Hindu Kush before any of these populations formed.

Pashtun AASI = Bronze Age Hindu Kush AASI
Punjabi/Sindhi AASI = Indus Valley AASI

They are not the same layer.
Pashtuns do not carry the Indus Valley ancestry that Punjabis/Sindhis have. Their AASI mixed with Iran_N/Chalcolithic in the highlands, not in the Indus plains.


Here’s what the data, not opinion, actually shows based on these sources:

1) Pashtun “South Asian” ancestry is ancient, Bronze Age, premixed with Iran_N long before Pashtun ethnogenesis.
2) Pashtuns show no recent Indo-Gangetic admixture.
3) Pakistani Pashtuns are among the most endogamous in the region.

Read Narasimhan et al. 2019 (Science):

Y-DNA does not prove 30–50% recent Indian mixing at all because autosomal DNA disproves it directly.

If you still want to argue otherwise, that’s fine, but I ask for one thing:

Bring a source. A paper, a study, A qpAdm model anything.
Something that explicitly says:
“Pashtuns derive 30–50% of their ancestry from recent Indo-Gangetic populations.”

I’ll leave it here until you bring a source.

Are you even reading what I'm saying? You misunderstood me, it's too frustrating.

I ALREADY SAID BEFORE IVC EXISTED IN CENTRAL ASIA BEFORE ARYANS ARRIVED, SUCH AS UZBEKISTAN. READ MY COMMENTS:






When I say 30-50% IVC, I MEAN THAT 30-50% OF THEIR CURRENT IVC VERY LIKELY COMES FROM INDICS. NOT ALL OF THEIR IVC COMES FROM THEM.
Example if pashtuns are 8% AASI, or 24% IVC. Then likely 7-12% of IVC that comes from indics, and their remaining were ALREADY PRESENT IN PASHTUNS. PASHTUNS ALWAYS HAD A CHUNK OF IVC, THEY LIKELY WERE MORE WEST EURASIAN BEFORE MIXING WITH INDICS.



"“The AASI ancestry in Hindu Kush groups reflects deep, pre-Bronze Age layers not attributable to recent South Asian admixture.”


Can you show me this quote? I didn't find it in section S3, nor in your link. Can you take a screenshot of that?

I even tried to look for you quote in the supplementary, I didn't ever see it. Nor did I see it in the link paper. Quite on the contrary, Narasimhan rather shown the AASI to been a subcontinental component that was rather spread by iranian neolithic pastoralists.




Yes, y-dna is different from autosomal. That doesn't mean now pashtuns have 0% indic ancestry at all, your study never claimed that either.



I never said ALL OF THE AASI IN CENTRAL ASIA WAS BROUGHT BY "ETHNIC INDIANS", I SAID THAT AASI IN BRONZE AGE CENTRAL ASIA WAS BROUGHT BY IVC OR OTHER WEST EURASIAN PASTORALISTS MOVING NORTH. WE DON'T SEE TUTKAUL HAVE ANY AASI, WHICH IS MUCH OLDER THAN BRONZE AGE. I SAID THAT PASHTUNS MOVED TO SULEIMAN MOUNTAINS, AND ADDED EVEN MORE IVC. I ALSO POSTED A STUDY OF TAJIKISTANIS ACCORDING TO DATES ADDING MORE SOUTH ASIAN IN 1000 CE. EVEN THE HISTORIC SWAT SAMPLES SHOWS INCREASED SOUTH ASIAN, COMPARED TO IRON AGE SWAT LOCALS. WHY IS THIS SO IMPOSSIBLE????


Swat Pashtuns already have a large amount of additional IVC compared to Suleiman mountains pashtuns, when they historically moved from Kabul to Swat. That happened in 1500s. That itself already destroys YOUR CLAIM that pashtuns never mixed additionally with indics "recently", when clearly can see THEY DID. The historic swat samples would been what pashtuns mixed with, and theyre closest to punjabis. Pashtuns in khyber often have local ghandaran haplogroups too not found in Suleiman mountains, but I guess that wouldn't mean they mixed with local indics at all, huh?

[Negah]

Are you even reading what I'm saying? You misunderstood me, it's too frustrating.

I ALREADY SAID BEFORE IVC EXISTED IN CENTRAL ASIA BEFORE ARYANS ARRIVED, SUCH AS UZBEKISTAN. READ MY COMMENTS:






When I say 30-50% IVC, I MEAN THAT 30-50% OF THEIR CURRENT IVC VERY LIKELY COMES FROM INDICS. NOT ALL OF THEIR IVC COMES FROM THEM.
Example if pashtuns are 8% AASI, or 24% IVC. Then likely 7-12% of IVC that comes from indics, and their remaining were ALREADY PRESENT IN PASHTUNS. PASHTUNS ALWAYS HAD A CHUNK OF IVC, THEY LIKELY WERE MORE WEST EURASIAN BEFORE MIXING WITH INDICS.



"“The AASI ancestry in Hindu Kush groups reflects deep, pre-Bronze Age layers not attributable to recent South Asian admixture.”


Can you show me this quote? I didn't find it in section S3, nor in your link. Can you take a screenshot of that?

I even tried to look for you quote in the supplementary, I didn't ever see it. Nor did I see it in the link paper. Quite on the contrary, Narasimhan rather shown the AASI to been a subcontinental component that was rather spread by iranian neolithic pastoralists.




Yes, y-dna is different from autosomal. That doesn't mean now pashtuns have 0% indic ancestry at all, your study never claimed that either.



I never said ALL OF THE AASI IN CENTRAL ASIA WAS BROUGHT BY "ETHNIC INDIANS", I SAID THAT AASI IN BRONZE AGE CENTRAL ASIA WAS BROUGHT BY IVC OR OTHER WEST EURASIAN PASTORALISTS MOVING NORTH. WE DON'T SEE TUTKAUL HAVE ANY AASI, WHICH IS MUCH OLDER THAN BRONZE AGE. I SAID THAT PASHTUNS MOVED TO SULEIMAN MOUNTAINS, AND ADDED EVEN MORE IVC. I ALSO POSTED A STUDY OF TAJIKISTANIS ACCORDING TO DATES ADDING MORE SOUTH ASIAN IN 1000 CE. EVEN THE HISTORIC SWAT SAMPLES SHOWS INCREASED SOUTH ASIAN, COMPARED TO IRON AGE SWAT LOCALS. WHY IS THIS SO IMPOSSIBLE????


Swat Pashtuns already have a large amount of additional IVC compared to Suleiman mountains pashtuns, when they historically moved from Kabul to Swat. That happened in 1500s. That itself already destroys YOUR CLAIM that pashtuns never mixed additionally with indics "recently", when clearly can see THEY DID. The historic swat samples would been what pashtuns mixed with, and theyre closest to punjabis. Pashtuns in khyber often have local ghandaran haplogroups too not found in Suleiman mountains, but I guess that wouldn't mean they mixed with local indics at all, huh?

My friend, you’re still arguing with a version of my position that I’ve never actually held. What is going on, my man? You are a very articulate person. Not sure what is going on here. I really thought we were about to have a great exchange of ideas and something profound and enlightening.

You keep repeating that “Pashtuns always had AASI, but then added 30–50% IVC from Indics recently” and you’re treating that as if it’s just obviously true because of some haplogroups and historical presence of Indians in the Suleiman mountains. Genetics is a science based on facts, not speculation based on historical narrative.

You have two problems:

1) You’re not engaging with the autosomal models. You need to.
2) You’re using Y-DNA + history to argue for quantitative autosomal claim of $\text{30–50\%}$ that the genome-wide data simply do not support and reject.

You need to have an autosomal model that shows Pashtuns = [Iran Plateau / Steppe / Ancient AASI] + big chunk of recent Indo-Gangetic source
With decent fit (qpAdm, DATES, etc.).

You wrote: “He only said the AASI in Hindukush is native there, and didn't come recently from example Pakistan like 1000–2000 years ago? Not that it's different.”

That is the whole point, my man.
That’s exactly the point.

Narasimhan’s whole setup is:
Ancient AASI (not recent) + Iran\_N/ChL mixed in the highlands. Then Steppe\_MLBA arrives and fuses with that.

Present-day Pashtuns, i.e., your people, are modeled as Iran-related + Steppe\_MLBA + that ancient AASI highland component (not from India)

When he tries to test recent Indo-Gangetic sources, those qpAdm fits break or are unnecessary. That is the key. If your “30–50% from Indics” story was right, then what do you think he should see? cause it's not there.

Here models Pashtuns as Iran-related + Steppe\_MLBA + ancient AASI-like South Asian hunter-gatherer component.

Then you say: “Swat Pashtuns already have a large amount of additional IVC compared to Suleiman Pashtuns … That itself already destroys YOUR CLAIM that Pashtuns never mixed additionally with Indics ‘recently’…”

No, that shows something much more modest and completely compatible with what I said, specially to Aviccena, which is that Different Pashtun groups on different frontiers picked up local shifts (Swat, some KPK zones, etc.). That does not imply a pan-Pashtun 30–50% Indic acquisition in the medieval period. Regional drift and local admixture absolutely. Pashtuns occupy a vast region, and I said they are diverse and not stable. Finally, your “huge segment of their IVC is late Indic from Suleiman” it is not demonstrated.

The real disagreement between us seems to be this.

You: A big chunk of Pashtun IVC / AASI comes from recent Indic mixing in the Suleiman mountains, which has 30–50% of their South Asian component; this is inferred from Y-DNA + linguistics + textual history.

Me: The bulk of Pashtun South Asian ancestry is ancient, formed via highland AASI + Iran\_N/ChL + Steppe\_MLBA, as modeled in Narasimhan-style qpAdm. Later Indian contacts may add local shifts, but not a massive, pan-ethnic 30–50% Indo-Gangetic input.

What I need for proof is to see a paper that says something like “Pashtuns derive a substantial portion of their ancestry from recent Indo-Gangetic groups (e.g., Punjabis/Sindhis) in the last 1–2 millennia.”

Otherwise, let's change the conversation and explore other things. You know where I stand, and I know yours. We can discuss other things.

[thisismyaccount]

My friend, you’re still arguing with a version of my position that I’ve never actually held. What is going on, my man? You are a very articulate person. Not sure what is going on here. I really thought we were about to have a great exchange of ideas and something profound and enlightening.

You keep repeating that “Pashtuns always had AASI, but then added 30–50% IVC from Indics recently” and you’re treating that as if it’s just obviously true because of some haplogroups and historical presence of Indians in the Suleiman mountains. Genetics is a science based on facts, not speculation based on historical narrative.

You have two problems:

1) You’re not engaging with the autosomal models. You need to.
2) You’re using Y-DNA + history to argue for quantitative autosomal claim of $\text{30–50\%}$ that the genome-wide data simply do not support and reject.

You need to have an autosomal model that shows Pashtuns = [Iran Plateau / Steppe / Ancient AASI] + big chunk of recent Indo-Gangetic source
With decent fit (qpAdm, DATES, etc.).

You wrote: “He only said the AASI in Hindukush is native there, and didn't come recently from example Pakistan like 1000–2000 years ago? Not that it's different.”

That is the whole point, my man.
That’s exactly the point.

Narasimhan’s whole setup is:
Ancient AASI (not recent) + Iran\_N/ChL mixed in the highlands. Then Steppe\_MLBA arrives and fuses with that.

Present-day Pashtuns, i.e., your people, are modeled as Iran-related + Steppe\_MLBA + that ancient AASI highland component (not from India)

When he tries to test recent Indo-Gangetic sources, those qpAdm fits break or are unnecessary. That is the key. If your “30–50% from Indics” story was right, then what do you think he should see? cause it's not there.

Here models Pashtuns as Iran-related + Steppe\_MLBA + ancient AASI-like South Asian hunter-gatherer component.

Then you say: “Swat Pashtuns already have a large amount of additional IVC compared to Suleiman Pashtuns … That itself already destroys YOUR CLAIM that Pashtuns never mixed additionally with Indics ‘recently’…”

No, that shows something much more modest and completely compatible with what I said, speicallay Aviccena, which is that Different Pashtun groups on different frontiers picked up local shifts (Swat, some KPK zones, etc.). That does not imply a pan-Pashtun 30–50% Indic acquisition in the medieval period. Regional drift and local admixture absolutely. Pashtuns occupy a vast region, and I said they are diverse and not stable. Finally, your “huge segment of their IVC is late Indic from Suleiman” it is not demonstrated.

The real disagreement between us seems to be this.

You: A big chunk of Pashtun IVC / AASI comes from recent Indic mixing in the Suleiman mountains, which has 30–50% of their South Asian component; this is inferred from Y-DNA + linguistics + textual history.

Me: The bulk of Pashtun South Asian ancestry is ancient, formed via highland AASI + Iran\_N/ChL + Steppe\_MLBA, as modeled in Narasimhan-style qpAdm. Later Indian contacts may add local shifts, but not a massive, pan-ethnic 30–50% Indo-Gangetic input.

What I need for proof is to see a paper that says something like “Pashtuns derive a substantial portion of their ancestry from recent Indo-Gangetic groups (e.g., Punjabis/Sindhis) in the last 1–2 millennia.”

Otherwise, let's change the conversation and explore other things. You know where I stand, and I know yours. We can discuss other things.

"Ancient AASI (not recent) + Iran\_N/ChL mixed in the highlands. Then Steppe\_MLBA arrives and fuses with that"

When does he say that? I never heard him say anywhere that AASI existed In the mountains prior iranian Neolithics arriving. Show me.




"Here models Pashtuns as Iran-related + Steppe\_MLBA + ancient AASI-like South Asian hunter-gatherer component."


If you meant "here" as he, why do you mention that? Narasimhan never mentioned pashtuns specifically or modelled them?



"No, that shows something much more modest and completely compatible with what I said, speicallay Aviccena, which is that Different Pashtun groups on different frontiers picked up local shifts (Swat, some KPK zones, etc.). That does not imply a pan-Pashtun 30–50% Indic acquisition in the medieval period. Regional drift and local admixture absolutely. Pashtuns occupy a vast region, and I said they are diverse and not stable. Finally, your “huge segment of their IVC is late Indic from Suleiman” it is not demonstrated."



Pashtuns are diverse, because they moved from Suleiman mountains to elsewhere. Otherwise mountain pashtuns have a small cluster.

When I say 30-50%, I mean just their IVC being 30-50% from indics. Not being actually 30-50% indic. Indics just have far more IVC admix than pashtuns do. So pashtuns being example 24% IVC and receiving example 8% from indics, means this 8% IVC translates to 12,5% Indic ancestry. Even though 8 out of 24 makes up to 33%. That's what I meant.


Dude, swat pashtuns very clearly shows founder effect in KPK. They do that, because a group of yusufzai pashtuns moved from Zabul, through Kabul, into Swat. Swat yusufzais had no significant presence prior the 1500s.

https://x.com/Afghan_DNA/status/1941795639488061806



Theyre the pashtun tribes with THE MOST IRANIC YAZ PASHTUN R1A CLADE. While this clade existing in other sarban tribes in Suleiman mountains, not equally as high amongst them. Yusufzais also have little amount of L clades. YES, they moved into KPK recently, the founder effect is clearly showing.


Yes, all pashtuns were in Suleiman mountains and perhaps part of Ghazni in 600-1000 CE, as WE SEE BY HISTORY. You can't deny historical accounts.



I keep saying VERY LIKELY (not 100%) mountain pashtuns have mixed with indics, because they would come from a region that rather resemble eastern shifted herat tajiks in north hazarajat. The locals in Suleiman mountains, based on the pre-aryan archeological sites rather related more to Indus than sistan, found in the mountains + the fact we know the pre-pashtun locals were indians and that some still live there today and near the mountains (they score like punjabis), they would been like Indus indics. Similiar to some of the swat Iron Age samples in the slopes of hindukush, who also are very close to punjabis.

Then compare modern pashtuns with those punjabis. Pashtuns are by far closer to tajiks than punjabis/indics, who would been the natives of Suleiman mountains. Pashtuns are slightly more south asian shifted compared to these tajiks, so it's obvious as day and night they mixed with them significant (probably derives 20% of their ancestry from them) based on the haplogroups. Watch, 2 mehsuds from Waziristan and another pashtun from Chaman, Balochistan. All 3 from Suleiman mountains.







https://x.com/vicayana/status/1916221824356004180. (9% AASI + 38% andronovo and rest BMAC + iran N)



The seraikis and other local indians haven't been modelled on qpadm, only seen them on g25. But I can see from their models, theyre just like other south punjabis and sindhis, such as sindhi hindus.

So I can compare these pashtuns with Sindhi hindus and aroras:


https://x.com/vicayana/status/1745406260433076589

https://gujjarancestry.substack.com/...is-genetically




Add khatris too:

https://www.reddit.com/r/SouthAsianA...ri_qpadm_runs/


The sindhis and punjabis are on average 22% AASI, while mountain pashtuns are 8-10% AASI.



There can't be such a big difference between what would been locals of Suleiman mountains, and present mountain pashtuns, if pashtuns didn't come from elsewhere. And as I've already shown, pashtuns by historical account, even seemingly linguistics, and by DNA, migrated to the mountains more than 1000 years ago. But since theyre slightly south asian shifted compared to Khorosan tajiks, it's not crazy to say at all they mixed with the locals THEY ASSIMILATED.


Man, i WOULD LOVED to post ancient DNA samples, but we have literally almost none in Afghanistan. Only 2-3 in the north. We can only go by guesswork, which at this point doesn't seem unreliable. We have haplogroups, the stark difference between pashtuns and mountain indics, compared to tajiks far north in northwest Afghanistan, the historical attestations. I don't think that's nothing?



Dude, there's not a paper FOR EVERY DAMN THING. You can use formal tools, use historical attestations to MAKE YOUR OWN CONCLUSIONS. However you can't keep ignoring how pashtuns have Suleiman mountain indic clades.




"Narasimhan’s whole setup is:
Ancient AASI (not recent) + Iran\_N/ChL mixed in the highlands. Then Steppe\_MLBA arrives and fuses with that.

Present-day Pashtuns, i.e., your people, are modeled as Iran-related + Steppe\_MLBA + that ancient AASI highland component (not from India)"


He modelled swat samples like that. Pashtuns are irrelevant to swat samples, man. Idk how this argument is negating anything at all?

Pashtuns being modeled as BMAC + indian tribal + andronovo means nothing, everyone is a mix of iran N, Anatolian, EHG, CHG, AASI in South Asia and Central Asia before arrival of turks. They can be modeled like that, and still have indic ancestry. It just means pashtuns were slightly more ANF shifted, before mixing with indics.


How much indic admix exactly? Idk, but definitely some admixture, since they have Suleiman mountain indic clades. They literally live in Indian territory, man, they would have absorbed a good chunk. We see the same for yusufzai pashtuns, compared to mountain pashtuns. Although probably not as much indic ancestry in mountain pashtuns.

Unfortunately we have no actual pre-turkic sample from anywhere in north Afghanistan, so I can't model pashtuns with said sample without any turkic ancestry + indic



Yes, no study support my theories. No studies reject them either, as they don't make much comment on it. Not every single damn thing is covered by a study paper.

[Peterski]

What it all comes down to is, there were catastrophes which interrupted the timeline, and calendars have been manipulated, and most information is controlled by "scholars" and "universities," (and corporations).

The confusion means we have to be very sceptical. "heliocentric theory" "evolution theory" "paleolithic age" "Yamnaya theory"

Wandal is simply a "flat-Earther" who writes utter nonsense which is against all of modern science.

[Avicenna]

"Ancient AASI (not recent) + Iran\_N/ChL mixed in the highlands. Then Steppe\_MLBA arrives and fuses with that"

When does he say that? I never heard him say anywhere that AASI existed In the mountains prior iranian Neolithics arriving. Show me.




"Here models Pashtuns as Iran-related + Steppe\_MLBA + ancient AASI-like South Asian hunter-gatherer component."


If you meant "here" as he, why do you mention that? Narasimhan never mentioned pashtuns specifically or modelled them?



"No, that shows something much more modest and completely compatible with what I said, speicallay Aviccena, which is that Different Pashtun groups on different frontiers picked up local shifts (Swat, some KPK zones, etc.). That does not imply a pan-Pashtun 30–50% Indic acquisition in the medieval period. Regional drift and local admixture absolutely. Pashtuns occupy a vast region, and I said they are diverse and not stable. Finally, your “huge segment of their IVC is late Indic from Suleiman” it is not demonstrated."



Pashtuns are diverse, because they moved from Suleiman mountains to elsewhere. Otherwise mountain pashtuns have a small cluster.

When I say 30-50%, I mean just their IVC being 30-50% from indics. Not being actually 30-50% indic. Indics just have far more IVC admix than pashtuns do. So pashtuns being example 24% IVC and receiving example 8% from indics, means this 8% IVC translates to 12,5% Indic ancestry. Even though 8 out of 24 makes up to 33%. That's what I meant.


Dude, swat pashtuns very clearly shows founder effect in KPK. They do that, because a group of yusufzai pashtuns moved from Zabul, through Kabul, into Swat. Swat yusufzais had no significant presence prior the 1500s.

https://x.com/Afghan_DNA/status/1941795639488061806



Theyre the pashtun tribes with THE MOST IRANIC YAZ PASHTUN R1A CLADE. While this clade existing in other sarban tribes in Suleiman mountains, not equally as high amongst them. Yusufzais also have little amount of L clades. YES, they moved into KPK recently, the founder effect is clearly showing.


Yes, all pashtuns were in Suleiman mountains and perhaps part of Ghazni in 600-1000 CE, as WE SEE BY HISTORY. You can't deny historical accounts.



I keep saying VERY LIKELY (not 100%) mountain pashtuns have mixed with indics, because they would come from a region that rather resemble eastern shifted herat tajiks in north hazarajat. The locals in Suleiman mountains, based on the pre-aryan archeological sites rather related more to Indus than sistan, found in the mountains + the fact we know the pre-pashtun locals were indians and that some still live there today and near the mountains (they score like punjabis), they would been like Indus indics. Similiar to some of the swat Iron Age samples in the slopes of hindukush, who also are very close to punjabis.

Then compare modern pashtuns with those punjabis. Pashtuns are by far closer to tajiks than punjabis/indics, who would been the natives of Suleiman mountains. Pashtuns are slightly more south asian shifted compared to these tajiks, so it's obvious as day and night they mixed with them significant (probably derives 20% of their ancestry from them) based on the haplogroups. Watch, 2 mehsuds from Waziristan and another pashtun from Chaman, Balochistan. All 3 from Suleiman mountains.







https://x.com/vicayana/status/1916221824356004180. (9% AASI + 38% andronovo and rest BMAC + iran N)



The seraikis and other local indians haven't been modelled on qpadm, only seen them on g25. But I can see from their models, theyre just like other south punjabis and sindhis, such as sindhi hindus.

So I can compare these pashtuns with Sindhi hindus and aroras:


https://x.com/vicayana/status/1745406260433076589

https://gujjarancestry.substack.com/...is-genetically




Add khatris too:

https://www.reddit.com/r/SouthAsianA...ri_qpadm_runs/


The sindhis and punjabis are on average 22% AASI, while mountain pashtuns are 8-10% AASI.



There can't be such a big difference between what would been locals of Suleiman mountains, and present mountain pashtuns, if pashtuns didn't come from elsewhere. And as I've already shown, pashtuns by historical account, even seemingly linguistics, and by DNA, migrated to the mountains more than 1000 years ago. But since theyre slightly south asian shifted compared to Khorosan tajiks, it's not crazy to say at all they mixed with the locals THEY ASSIMILATED.


Man, i WOULD LOVED to post ancient DNA samples, but we have literally almost none in Afghanistan. Only 2-3 in the north. We can only go by guesswork, which at this point doesn't seem unreliable. We have haplogroups, the stark difference between pashtuns and mountain indics, compared to tajiks far north in northwest Afghanistan, the historical attestations. I don't think that's nothing?



Dude, there's not a paper FOR EVERY DAMN THING. You can use formal tools, use historical attestations to MAKE YOUR OWN CONCLUSIONS. However you can't keep ignoring how pashtuns have Suleiman mountain indic clades.




"Narasimhan’s whole setup is:
Ancient AASI (not recent) + Iran\_N/ChL mixed in the highlands. Then Steppe\_MLBA arrives and fuses with that.

Present-day Pashtuns, i.e., your people, are modeled as Iran-related + Steppe\_MLBA + that ancient AASI highland component (not from India)"


He modelled swat samples like that. Pashtuns are irrelevant to swat samples, man. Idk how this argument is negating anything at all?

Pashtuns being modeled as BMAC + indian tribal + andronovo means nothing, everyone is a mix of iran N, Anatolian, EHG, CHG, AASI in South Asia and Central Asia before arrival of turks. They can be modeled like that, and still have indic ancestry. It just means pashtuns were slightly more ANF shifted, before mixing with indics.


How much indic admix exactly? Idk, but definitely some admixture, since they have Suleiman mountain indic clades. They literally live in Indian territory, man, they would have absorbed a good chunk. We see the same for yusufzai pashtuns, compared to mountain pashtuns. Although probably not as much indic ancestry in mountain pashtuns.

Unfortunately we have no actual pre-turkic sample from anywhere in north Afghanistan, so I can't model pashtuns with said sample without any turkic ancestry + indic



Yes, no study support my theories. No studies reject them either, as they don't make much comment on it. Not every single damn thing is covered by a study paper.

Would you propose they mixed with these types of people?

https://www.instagram.com/reel/DO8nb...xpMWNsdDczZg==

[thisismyaccount]

Would you propose they mixed with these types of people?

https://www.instagram.com/reel/DO8nb...xpMWNsdDczZg==

Yes, I would assume so. Pashtuns would been seemingly more northern shifted, with probably far less south asian looking individuals prior

[Avicenna]

Yes, I would assume so. Pashtuns would been seemingly more northern shifted, with probably far less south asian looking individuals prior

This does actually make sense tbh, I would have imagined OG pashtuns to have looked like those darwaz badakshi Tajiks, absolutely zero south Asian or zagros heavy phenotypes among them, but moving towards the suleiman mountains caused them too assimilate the local hill populations ( partially obviously ) and increasing their IVC ancestry and absorbing some phenotypes from them. I think the way you phrased it to Negah made it seem like they are a hybrid between the two when in reality they only partially assimilated these folks.

Edit: even those darwaz badakshi Tajiks with zero south Asian looking people still genetically have some IVC and as a result AASI in their genome, so OG pashtuns would have probably been about 2-4% AASI, with assimilation of these local suleiman populations increasing their AASI to around 6-8%.

[Negah]

@thisismyaccount

My friend, before I go point-by-point, I need to lay out something very simple to you. After reading all your replies several times, I’ve identified three major weaknesses in your argument, and unless we fix these, the conversation will be painful for both of us.

I ask you kindly, please read this post, and please address all 3 issues I have identified; otherwise, the conversation will keep going in circles.

1) You keep treating my posts as “my personal views.”

None of what I wrote is “my opinion.”
Everything I posted comes from peer-reviewed genetic papers; they are from Narasimhan, Lazaridis, Skoglund, Reich lab, etc. I even quoted them directly with links to make it easy. So when you tell me “you’re wrong,” you’re actually disagreeing with the current scientific consensus that I am sharing with you, not me. I am not a geneticist, same as you. I am just stating what these professionals say.

If you think these scientists are wrong, then you need to show scientific counter-evidence. Telling me I’m wrong while ignoring the sources I posted is not a scientific argument.

2) You’re relying on tools that are not scientific (G25, HarappaWorld, GEDmatch, etc).

You cannot use G25, GEDmatch, HarappaWorld, 23andMe, Ancestry, MyHeritage “2% Chinese, 3% Samurai, 1% Swedish” type reports or PCA screenshots to talk about ancient ancestry.

Why? Because these are hobby tools. They are entertainment. Not science.
They cannot detect deep admixture, migration direction, or timing. They are not used in any peer-reviewed work.

Professional population genetics uses qpAdm, qpGraph, DATES, and f-statistics.
These are the only tools that detect ancient ancestry properly.

This is the same mistake our Kurdish friend (parents from Turkey, born in Germany) made years ago when he tried to show Kurds had “no South Asian ancestry” by misusing these calculators. Back then you told me he didn’t know what he was doing. You even laughed at how he was twisting hobby tools to fit his agenda.

Now you’re repeating the exact same error.

Please cite peer-reviewed papers where real geneticists use the right tools. Genetics is a science. It is based on fact, not assumption, not inference, not hobby calculators, not spreadsheets.

3) You use history to prove genetics, for instance, you’re using Hindu Shahi / Kabul Shahi history as if genetics uses medieval dynasties as data.

Geneticists do not use dynasties, chronicles, Biruni quotes, or medieval political history to reconstruct deep ancestry.
They use DNA and scientific toolkits. By its very nature, genetics (and archaeology) is conservative because they soley and only rely on evidence and facts. They cannot make historical inferences. Historians can, because they have multiple tools—texts, archaeology, linguistics, Genetics, numismatics, etc. Geneticists cannot.

We can talk history, but you cannot use history to prove genetic conclusions. That is simply not a correct or proper methodology.

And historically, the Hindu Shahi / Kabul Shahi dynasty has nothing to do with Pashtuns.
The Shahi domains were a mixture of:

1. Bactrian elements
2. Indo-Aryan / Prakrit-speaking elements
3. Central Asian elements

none of that = “Pashtuns.”

Modern Kabul has zero linguistic, cultural, or genealogical continuity with the Shahi court.
Kabul’s population was replaced many times: Ghaznavids, Ghurids, Mongols, Timurids, Mughals, Durranis, plus modern-era shifts. If you want, we can discuss it. But Today, Kabul is a Tajik-majority city with Hazara and Pashtun among many minority groups; it is nothing like medieval Kabul.

You cannot use “Hindu Shahi” as a genetic category. That is not how genetics works.

4) Stop misquoting me. I never said Pastun had no AASI. You do, and you have a lot of it. It is a foundational part of your ancestry. That is a fact. What I said one more time and please only discuss this and stop misconstruing it. It is very simple, based on the studies that cite not my view but the studies that are reputable, peer-reviewed, and based on solid scientific foundations, not myth, folklore, nationalistic narratives, colorism, or political persuasion.

Pashtuns do have AASI — and quite a lot of it.
So do Tajiks.
So do Iranians.
So do Central Asians around the old BMAC zone.

This is not controversial.
This is basic population genetics.

The real question is:
Is that AASI recent (medieval Indian) or ancient (pre-IVC highlander AASI)?

The peer-reviewed papers say it very clearly:

Pashtun AASI = ancient highland South Asian ancestry
Not medieval Punjabi/Sindhi input.

This is exactly what Narasimhan et al. show. I quoted the line before:

“The AASI ancestry in Hindu Kush groups reflects deep, pre-Bronze Age layers not attributable to recent South Asian admixture.”

Narasimhan et al., 2019, Supplement S3

https://pmc.ncbi.nlm.nih.gov/articles/PMC6822619/

https://reich.hms.harvard.edu/sites/...Supplement.pdf

So Avicenna is wrong when he uses “steppe” to explain being lighter and uses “Indians” to explain being darker. That’s not science. That’s phenotype guessing.

And you are wrong when you say Pashtuns got 30–50% AASI from medieval Punjabis.
The real data simply do not support that.

Pashtuns have AASI, yes.
But the source of most it ( not all of it) is ancient — older than “India” as a concept, older than Indo-Aryans, older than the Shahi dynasties, older than the modern ethnic map.

This is why Pashtuns sit genetically:

closer to Iranians, Kurds, Tajiks
not
Punjab/Sindh.

The reality is that you and I are cousins of Punjabis, but mainly an ancient cousin, not a medieval one.

The Pashtuns drift south on PCA because of their ancient AASI, not because of a massive medieval Indian mixture.

That’s all I’ve been saying from the beginning.

[Negah]

Yes, I would assume so. Pashtuns would been seemingly more northern shifted, with probably far less south asian looking individuals prior

Genetics is not phenotype. That is voodoo science. Skin color is determined by a relatively small number of genes. The genes for light skin like the SLC24A5 allele associated with Steppe populations.