Magyar Origins (Second Edition)

A 21st Century Look at the Origins of

Ancient Hungarians

Hungarian DNA Is Not From North Asia

(Internet Explorer 10 or 11 please view in compatibility mode)

by Frank Sandor   January 1, 2013

In my book Magyar Origins : A 21st Century Look at the Origins of Ancient Hungarians, I discuss how most Hungarian males belong to the genetic haplogroup called R1a1a.  Since the book was written the percentage of the population estimated as being R1a1a has been revised by geneticists from 60% to 25%.  While this is a large change R1a1a still remains the largest DNA group in Hungary.  And when known resident populations like Germanic DNA and Slavic DNA are factored out, the percentage of R1a1a DNA comprising the Hungarian descendents of the original Magyar migration in the 9th century, rises again to approximately 50%.  The question that must be addressed is how the more recent DNA findings affect Chapter 1 and any reference to DNA that I make in my book.   The answer, to put it simply, is that the more recent DNA findings further supports my statement in the book that Hungarians are from the Hindu Kush region.

I make this claim not only because R1a1a and the vast majority of DNA types found in Hungary are found in the Hindu Kush region, but also because certain anomalous DNA results can only be explained by considering a Hungarian presence in the Hindu Kush region, (for example, the lack of North Asian DNA amongst Hungarians except in the warrior caste and the discovery of an Indush Valley/Pakistani gene in a small percentage of Hungarian males). 

In my book I predicted that the warrior caste in Hungary would have a different DNA haplogroup than the majority of the Hungarian population.  At the time I knew of only two 9th century warriors to have undergone genetic testing and both belonged to the haplogroup N1c1 yet the DNA sampling at the time showed a majority of Hungarians to be R1a1a.  Since I wrote the book it has been confirmed by Istvan Rasko, Director of the Institute of Genetics of the Hungarian Academy of Sciences, that the warrior caste in the 9th century was predominantly N1c1 and that the warrior caste contributed very little DNA to the ‘common person’ population of Hungary (Raskó, István, et al, Analysis of Paternal Genetic Relationship of Ancient Hungarian, Modern Hungarian and Modern Szekler Populations, Budapest, 2009).  In other words the DNA haplogroup of the warriors was N1c1 but that haplogroup is almost nonexistent in the main population.  This has lead to the incorrect conclusion by some that only the N1c1 warriors were the settling population and that the bulk of Hungarians must have pre-existed in Hungary.  Otherwise the common population would have obtained the N1c1 gene. 


This logic violates what we know about the spread of languages.  If the small number of warriors were the only ones speaking Magyar they would have been absorbed by the local population and taken on their language.  History shows that when a small group speaking a different language imposes its language on the masses, the imposed language will usually be for administrative purposes but it never takes over the language spoken at social gatherings. 

In South America the Spanish minority were able to spread their language to the non Spanish masses through the use of religion.  As the Spanish spread their religion they spread their language.  But in Hungary no such avenue existed for Magyar.  There were no missions established to teach the Magyar language and religion to a conquered people.  In fact it was the other way around.  Missions were established to rid Hungarians of their religion and language.  Latin effectively replaced Hungarian writing through this means and became the administrative language of the country.  King Stephen I actively fought local populations to replace Hungarian with Latin.  This is not exactly the actions of a Magyar ruling class imposing their imported language on the masses.

This is repeated over and over again in the homes of Hungarian immigrants.  No matter how hard they try to have their children speak Hungarian at home, it is little more than an administrative language of the parents.  Outside the home the children, out of social necessity, speak the local language.  The children of the Magyar warriors would have experienced this exact same problem if they were the language minority.

There are also those who believe Hungarian was the language of the masses that existed in Hungary before the Magyar warriors arrived, completely ignoring all cultural and linguistic evidence.  They forget that Hungarian is a Uralic language and that there is no evidence to support the idea that the Finns or Estonians originated in Hungary. That is what would be needed to explain the Uralic languages if you don’t believe Hungarian was a language brought into the country in the 9th century. 

To assume the warriors did not live in the same culture as the common people because they did not share DNA is an incorrect assumption.  As I state in my book, Hungarians in the 9th century practiced a form of Vedic-Hinduism.  Amongst the Hindi who also have a religion based on Vedic-Hinduism, there are five castes.  Of those the warrior and priest castes are the upper castes and are prohibited from ‘mixing’ with the merchant/agricultural and labour castes.  The lowest ‘untouchable’ caste was not even allowed to be on the same street as warriors.  Mixing of the genes without priest approval could result in ostracization or even death.  The result is genetic isolation of the warrior caste from the common population.  Even among those Hindu castes that are allowed to ‘mix’, the exchange of DNA is less than one percent per year (Zerjal et al, National Center for Biotechnology Information, Hum Genet Manuscript PMC2590678, 2008).  The same genetic isolation of the warriors that was found in Hungary is consistent with Vedic-Hinduism and parallels what is found in Pakistan and India.  The small number of genetically isolated warriors and the large number of common people, also verifies the concept I introduce in my book that a majority of 9th century Magyars were farmers and not warriors. 

Another thing that has changed since the writing of my book is the discovery of the Z280 genetic marker in Hungarians belonging to the R1a1a group.  At the time of its discovery, because of the study’s heavy contribution of Hungarians, it was believed this marker only existed in Hungary.  Since that time this marker has been shown to be common throughout Central Asia including having recently been discovered in Kazakhstan and Uzbekistan.  The proximity of Uzbekistan to Pakistan suggests it is only a matter of further study before the Z280 marker is also found there as no geographic barrier between the two countries until you get to the Hindu Kush Mountains near the Indian border.  (The distance between Pakistan and Uzbekistan is about the same distance as a return trip from Washington, DC to New York.)

In 2009 geneticist Spencer Wells concluded that Hungarians did indeed come from Central Asia  (Wells, Spenser, Genetic Analysis Reconfirms That Hungarians are Europe’s Asians,, Oct 1, 2009). For Wells, Central Asia includes the region from the Urals to the Hindu Kush in Pakistan.  In my book I include the Hindu Kush as part of South Asia because Pakistan is part of the Indian subcontinent.  Wells says the exact location of genetic origin for Hungarians and the Z280 marker is still undetermined, he just narrows it down to Central Asia.  However, the Y-DNA group L (M20) has been found in a very small percentage of Hungarian men and it is considered an Indush Valley / Pakistani marker, indicating ancestry from Pakistan or Northwest India.  

The FamilyTreeDNA Hungarian Magyar Y-DNA Project provides a list of the genetic types found in their study to date.  An important thing to note is that this site, unlike most, admits that DNA testing is still in its infancy and the place of origin for these DNA groups is mostly guess work. 

The Y-DNA types listed in the FamilyTreeDNA Hungarian Magyar Y-DNA Project, as of Dec 2010, include:

(The haplogroup breakdown (N=203))

Genetic type

Percentage of study with this type


Found in Hindu Kush region



(with Z280 marker) - Found from the Balkans to the Hindu Kush


(with Z93 marker) - Traditional theories claim this is a Turkish marker but it is also an Indo-Iranian marker and is found in east Iran and India.



19.5 %

This most likely the original population of the Carpathian Basin, and is consistent with Celtic or Germanic Ancestry.



13 %

Consistent with Slavic ancestry




Consistent with Germanic ancestry




Illyrian, Sudanese and Semitic DNA all three types found in Hungary.  FamilytreeDNA has no explanation how Sudanese DNA entered Hungary but all three subgroups of the DNA group are consistent with the Greek conquest of Persia and Northern India.




It is believed this DNA entered Hungary with the Kabars or Alans




A Fertile Crescent gene.  It originated from Anatolia to the Persian Gulf, and is consistent with the Greek conquest of Persia and Northern India.




North Eurasian.  N1c1 was the DNA of the 9th century warrior caste.




Inner Asian group. 




Indicates German ancestry




Mediterranean ancestry, consistent with the Greek conquest of Persia and Northern India.




Xiōngnu or Avar in origin.  I doubt it came from the Mongolian attack on Hungary as that incident left an estimated 50% of the population dead and it is unlikely Mongolians were leaving their sons behind to raise families.




Indush Valley / Pakistani origin.  Conventional theories have no explanation how this marker made its way into the Hungarian population.


So what do these numbers tell us?  First keep in mind that these percentages are based on so few people that they can change significantly with the addition a just a few people to the study.  For instance larger studies have shown J2 to be 16% of the Hungarian population and as high as 22% in Szekely, Romania. 

J2 is a gene from the Fertile Crescent of the Middle East and the Mediterranean.  Dr. Rasko speculates that this gene may have been brought into Hungary during the Turkish occupation.  This is unlikely as Bulgaria is known to be home to the largest population of Ottoman Turks outside of Turkey and ethnic Turks make up only 10% of Bulgaria’s population.  The concentration Dr. Rasko found of J2 in Sekely is over twice as high as the total percentage of all Turkish decedents in Bulgaria.  Another inconsistency is that most Hungarians that are J2 are J2a which is common in Greek colonized areas such as the Hindu Kush region, the J2 in the Balkans is predominately J2b so migration from the Balkans is unlikely. 

Speculation of the Ottoman Turks as the source of J2 also ignores Hungarian history.  Dr Rasko sampled 97 people of Szekler origin and found 22% of them to be J2.  If the gene is of Turkish origin then one in four Szeklers are of Turkish descent.  Of all the regions that once formed part of Hungary, Szekely was almost completely untouched by the Turks.  A one in four Turkish origin in recent times would counter the Szeklers claim to be one of the most ‘pure’ Magyar villages still left.  They proudly boast that they are more Hungarian than the Hungarians, a claim that would have quickly been criticized by other villages if they thought Szekely was a Turkish town.  Szekely’s ‘purity’ was one of the reasons they were included in a study comparing modern and ancient DNA.  So, if the DNA of the Hungarian common people is 16 to 22% Mediterranean, and the Turks didn’t bring a majority of that gene or the other genes of Mediterranean origin like T with them, then the only other possibility is that Hungarians lived in a region rich with Mediterranean DNA.  In Asia this is not the Urals but is the Hindu Kush which was occupied by the Greeks in the 4th century BC.   

Knowing that Hungarians contain a much higher percentage of Mediterranean DNA than geneticists expected we can group Hungarian DNA into regions of origin.  We know that German populations were deliberately settled into the Carpathian Basin all the way back to King Stephen I as a means of exercising military control.  We also know that Slavic populations would have been native to the area so we can exclude these two groups from any migration study.  That would exclude 45.5% of this study or 92 people.  From the remaining DNA groups almost all are found in the Hindu Kush region with the exception of the DNA of Kabar origin, which is known to have entered Hungary as separate tribes with the Magyars in the 9th century. 

If we group Hungarian DNA by region of origin it paints a significantly different picture than what one would expect from a culture that would have originated near the Ural Mountains, or from what is found in Finns and Estonians who are mostly N3 and R1a1a.

Region of origin

DNA group

Total Percentage of FamilyTreeDNA Project

German, Slavic

R1b1b2, I2a2, I1, I2b1

19.5 + 13 + 10 + 3 = 45.5

Central Asian (up to the Hindu Kush)

R1a1 (mostly Z280 but some South Asian Z93 as well)


Mediterranean, North African

E1b1b1, J2, T

8 + 8 + 1 = 17

North Asian, Inner Asian

N, Q, C3

3.5 + 3 + .5 = 7








When German and Slavic DNA is removed from the FamilyTreeDNA study it demonstrates a stark contrast between Hungarians and the other Uralic speakers like Finns and Estonians.  Finns and Estonians are predominantly Central and North Asian genetic types.  Hungarians on the other hand, are predominantly Central Asian and Mediterranean in origin.  The lack of the North Asian N gene in Hungarians proves Hungarians were not part of the northern Uralic migration.  This is strong evidence supporting what I stated in my book that the Finns and Estonians migrated north while the Hungarians remained in the south near the Hindu Kush, a region conquered by the Greeks during the time of Alexander the Great.  By leaving before the Greeks arrived the Finns and Estonians did not have a chance to absorb the Mediterranean DNA.  Recent DNA research clearly proves that those Uralic speakers that went north acquired the North Asian genes and those that remained in the south acquired the Mediterranean genes, resulting in two completely different mixes of Uralic DNA. 

People who claim that the only “true Magyars” were the warriors will point out that the N group is a Uralic family DNA group and is common throughout Northern Europe.  The problem with this argument is that its supporters are only looking at the parent gene.  The majority of Hungarian warriors (72%) and the gene they passed on to the modern population is N1c1, without the N3 TAT mutation.  Only the Finns and Estonians predominately carry the N3 TAT mutation.

N3 in the Finns and N1c1 in the Hungarian warriors is consistent with what I state in my book, that the Finns broke from the Hungarians and migrated north to Siberia leaving the Hungarians in the Hindu Kush region.  From Siberia N3 was picked up and carried west into Finland and Estonia.  Hungarians remaining in the Hindu Kush would have acquired N1c1 with the Hun tribes that conquered that area much later, along with the Q and C genetic types.  This is also consistent with Hungarian legends claiming that at least some warriors were decedents of Huns.  Dr. Rasko speculates that the Hungarian warriors were possibly at one time also N3 and lost the mutation through genetic drift.  This suggestion offers no explanation on why N is common in Finns and Estonians but absent in Hungarians who are not of warrior ancestry.  Only 3.5 % of Hungarians today are N compared to 35% of Estonians and 60% of Finns.  The N1c1 gene in Hungarian warriors must have been acquired after the Finns separated or we would expect to see the same rate of transmission of the N gene in both Hungarians and Finns.  By migrating to Siberia before acquiring a social restriction preventing the gene from spreading, the Finns were free to absorb N3 throughout their entire population.  Hungarians on the other hand must have acquired the N1c1 gene after a social / religious restriction developed preventing the warrior caste from spreading it throughout the population.

Since the Hungarian common person of the 9th century did not carry the North Asian N gene, R1a1 must be the common genetic group between Hungarians, Estonians and Finns as the Z280 marker common in Hungarians is also common throughout the Ural Mountain Region and Z280 is now considered a Finno-Ugric group and not just a Hungarian group.  So to find the point of origin of all the Uralic speakers, regardless if Hungarians are 25% R1a1a or 60% R1a1a, you must look for a location that can account for that haplogroup in both the Hungarians and the other Uralic speakers.  This location must also not contain a high concentration of the N DNA group or Hungarians would have acquired it at the same rate as the Finns.  This rules out Siberia or the Ural Mountains as a point of origin for the Uralic languages. 

There is one school of thought that believes the Z280 marker actually originated in Eastern Europe, possibly Poland, from there the gene spread into Central Asia.  Another school of thought believes the spread of Z280 occurred in the opposite direction.  Regardless of where the first person with Z280 was born it ultimately ended up in the Hindu Kush region, and for Hungarians this was a significant event because that is where the foundation of their language and culture originated, and this is consistent with Hungarian legends claiming a Persian origin.

When the fact that Hungarians are a mix of Central Asian and Mediterranean DNA is combined with the linguistic and cultural evidence I present in my book, the only logical conclusion for the origins of the Uralic peoples is the Indian subcontinent, in the region of the Hindu Kush.

Additional information since publication (January 22, 2014):

Since this article was original written genetic research has made some advances.

According to the International Society of Genetic Research, Hungarian males, and the Magyar Asian warriors, who do belong to haplogroup N1C1 also carry the marker L1034. This marker is considered to be a northern Eurasian Ugric marker. It is found in Hungary, and from the Urals eastward including Mongolia, Kazakhstan, Uzbekistan, etc.  It is not found in the Baltics.  Proving conclusively that the migration of the Uralic speakers to the Baltics occurred much earlier than the migration of haplogroup N1C1 to Hungary and was in fact a separate migration.  

There is a school of thought that considers the Bashkirs to be the origins of the Magyar warriors.  The Bashkirs currently reside in the Ural region.  This at first glance is possible.  They share a similar language, and are the same genetically, consisting of both N1C1 L1034 and R1a1 Z280.  In the 12th century Hungarians were even called Bashkirs by Al-Garnati.  Unfortunately for this school of thought the Bashkirs settled in the Urals in the 9th century where they acquired both the genetic types and the Uralic language.  Prior to this the Bashkir spoke a Turkic language of which almost no traces remain.  Linguists believe that all of the similarities between Bashkir vocabulary and the Uralic languages are due to the fact the entire Bashkir language has been replaced by loan words, but the core of their language is still Altaic and not Uralic.  So it is impossible for the Bashkir to be the ancestors of the Magyars.

The Khanty-Mansi share the genetic type N1C1 L1034 as well as a language of similar origins.  It is far more likely the Khanty-Mansi share a common ancestor with the Magyar warriors than any suggested Bashkir connection.  It is commonly accepted that the Khanty-Mansi are also recent arrivals to their current home. Originally they lived much further south and west of their current location, only latter (around the 5th century AD) migrating north as a result of the Hun wars.  This final push northward for the Khanty-Mansi was just another step in a northward migration for this genetic type whose distant ancestors originated near the Sudan, taking them on a path through the Pamir Knot and the Hindu Kush.  If Magyar warriors share their genetic type with the Khanty-Mansi then it is logical to conclude that the Magyar warriors originated somewhere along this same migration route.

Click on Image to Enlarge


Eupedia Haplogroup J2

Eurasia 98, Sample sites for Dr Wells study


FamilyTreeDNA Hungarian Magyar Y-DNA Project

FamilyTreeDNA Hungarian Magyar Y-DNA Project  results

FamilyTreeDNA  India subcontinent DNA Project

FamilyTreeDNA N Y-DNA Haplogroup Project- Background

FamilyTreeDNA N1C1 Y-DNA Project

FamilyTreeDNA Afghan-Pakistani DNA Project

FamilyTreeDNA R1a1a and Subclades Y-DNA Project

FamilyTreeDNA Romany DNA project - Y-DNA Classic Chart


FamilyTreeDNA Brief Communication: New Y-Chromosome Binary Markers Improve Phylogenetic Resolution Within Haplogroup R1a1

Rasko et al, Analysis of paternal genetic relationship of ancient Hungarian, modern Hungarian and modern Szekler populations

UCLA Language Materials Project

Y-chromosomal insights into the genetic impact of the caste system in India. SEMARGL.ME  Z280 study

Vajda, Edward The Khanty and Mansi

Wells, Genetic Analysis Reconfirms That Hungarians are Europe’s Asians, 2009


Wikipedia Haplogroup N-M231 (Y-DNA)