I-Y16419 1300 BC Indigenous or Balkan migrants to Armenia?

I-Y16419 is found today in Italy, the Balkans, Eastern Europe, Turkey and the Caucasus, as well as a few outliers in England, Germany and Mexico. All these men descend from one man who lived 1300 BC. The 13,000 years preceding this man was a bottleneck, a single lineage from father to son to son (etc) with no other surviving siblings.

Most I2 subclades are European, but it is not so straightforward to explain the origin and subsequent migrations of Y16419. His sibling S6635 is found all over Europe and new sibling BY160929 found on FTDNA has just two samples, an Egyptian descending from Dawud Agha and a Reunion Islander.

Given the large geographic distribution of these siblings and the immense timescale involved, analysis of Y16419 subclade distribution is necessary to get a better idea of where the founder lived.

Was the man at the end of the bottleneck in 1300 BC living in Southeast Europe, Eastern Europe, Anatolia, the Caucasus, or elsewhere?

The The I-L596 FTDNA Project page contains many more samples than are shown on the YFull at this time.

I analyzed the STRs of these samples into the following subclades, in some cases splitting them further on the basis of shared mutations. (Country level detail)

Deepest diversity of Y16419 is with Armenian people and geographically, Anatolia and the Caucasus

Origin of Y16419 computed as NW Caucasus. Note this is based only on YFull data which is very scant. I will recompute later on full set of data.

Two theories I will consider here are that the ancestor was a Proto-Armenian living in eastern Anatolia/Armenia or that the ancestor was living in the Balkans and a member of a Paleo-Balkan / Proto-Armenian tribe whose progeny over the next centuries participated in the ethnogenesis of the Greeks, Albanians and Armenians.

Armenian / Eastern Anatolian Origin

If the ancestor in 1300 BC was living in eastern Anatolia or Armenia then the most troublesome regional distributions are in Italy and the Balkans along with the German, English and Mexican outliers. It is also far afield from the rest of I2 in Europe.

It would mean that the two Pontic Greek subclades were originally Armenian (or Proto-Armenian) Anatolians who adopted the Pontic Greek cultural identity after Greek colonization of the coasts. This could have happened via two Armenian men marrying into Pontic Greek families. In both Pontic Greek subclades A1143 and Y16418>BY4177 the Anatolian Armenian and Pontic Greek geographic distributions nearly overlap. It is worth noting that A1143 project contains three Greeks of non-specific origin while both Greek subclades contain Pontic Greeks.

An Armenian presence in Italy, Thrace and Macedonia could be due to Armenian religious refugees known as Paulicians, or a number of other waves of exodus.

Balkan Origin

There are a number of Indo-European speaking groups, starting with the Hittites, for which we have good archaeological and historic evidence that they migrated from the Balkans to Anatolia.

If Y16419 originally lived in the Balkans then the bulk of the successful lineages were formed by members who migrated in a similar fashion to eastern Anatolia and Armenia. In this case, the descendants currently living in Europe are the remnants of those that stayed behind rather than needing to be explained by more recent Armenian or Armenian turned Pontic Greek-diaspora.

It is generally assumed that the Hittites and Luwians came into Anatolia some time before 2000 BC. However the Y16419 diversification happened later (current estimate at time of writing 1300 BC), around the time of the late Bronze Age Collapse of 1200 BC. During what has been described as a systemic collapse of trade networks, many civilizations around the Mediterranean went into decline. Many centers of population and palatial sites were destroyed or abandoned. An informative book about this is 1177 BC: The Year Civilization Collapsed by Eric H Cline.

By Alexikoua - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=30160275

After the collapse of the Hittite Empire the political vacuum in central-western Anatolia was filled by the Phrygians.

A supposed Armenian migration from the Balkans is not well supported like the Phyrgians. But it appears that after the fall of Urartu around 600 BC, an Indo-European language with no prior written antecedents became spoken among a population of Hurro-Urartian langauges in the mountainous region of Eastern Anatolia (Clackson, James P. T. 2008. Classical Armenian. In: The languages of Asia Minor (ed. R. D. Woodard). P.125.). Perhaps the Indo-European nucleus of this group had accompanied the Phrygians from the Balkans and with them overthrew the Hittites.

The Graeco-Armenian theory posits that Greek and Armenian derived from a common ancestor language known as Graeco-Armenian. This theory is not generally accepted but the lack of a confirmed linguistic link  does not preclude the possibility of some kind of Paleo-Balkan Indo-European speaking tribe having participated, to varying degrees, at least genetically in the ethnogenesis of Albanians, Greeks, Armenians or peoples who spoke now extinct Indo-European languages of the Anatolian or Paleo-Balkan branches.

Are there similar genetic links between Armenian and Balkan subclades in other haplogroups with TMRCA 1300 BC?

Late Bronze Age Caucasus

The next state to emerge in this region after the fall of the Hittites in the 12th century BC is Urartu. During the 11th and 10th centuries BC, the kingdom eventually encompassed a region stretching from the Caucasus Mountains in the north, to the borders of northern Assyria and northern ancient Iran in the south, and controlled much of eastern Anatolia.

Though it's leaders spoke Urartian, it is unknown what language was spoken by all the peoples of Urartu at the time of the existence of the kingdom. However, there is linguistic evidence of contact between the proto-Armenian language and the Urartian language at an early date (sometime between the 3rd—2nd millennium BC), occurring prior to the formation of Urartu as a kingdom.

If any of the Georgian subclades of Y16419 end up having TMRCAs near 1300 BC it could mean that they entered Kartli in the centuries following the decline of the Trialeti Culture, possibly from Urartu.

Some questions that would shed further light:

  • Where do the English, German and Mexican outliers fall in the tree and when did their ancestors get to these places (in the Mexicans case, from where did they originate in Europe before colonizing Mexico?
  • How old is the TMRCA of the largely Ashkenazi BY2807 and where did he appear to have lived?
  • How old is the TMRCA of exclusively S/SE European subclades?
  • How old are the TMRCAs of exclusively Georgian subclades?

None of the few Georgian samples I have placed in subclades have any close matches. There are also at least four other Georgians who are so distant from other samples that I have left them off the tree. This includes two men from Khevsureti who share very rare STR markers (YCAII=19-20 and DYS464d=17) but have GD 27/111. This may indicate that many Georgians in Y16419 are survivors of old bottlenecks, some perhaps dating back to the original Y16419 man who entered the Caucasus.


Most I-Y16419 lineages are found today in eastern Anatolia and the Caucasus, from Turkey to Armenia, Iran, Azerbaijan, Georgia, North Ossetia, Kabardino-Balkaria, Karachay.

There are two reasonable theories regarding the origin of the Y16419 man who lived in 1300 BC.

  1. If he was born in the Balkans, then most of his surviving sons migrated to eastern Anatolia, Armenia, Georgia and Ossetia in the years after 1300 BC. Given that there are no surviving western Anatolian lineages it appears that all the migrants went to eastern Anatolia together.
  2. If he was born in eastern Anatolia or Armenia, then his descendants living in Europe today may be more recent immigrants from the Greek or Armenian diaspora.

SNP testing has not conclusively established the ages of the exclusively ethnic Georgian lineages (as it has for the Armenian ones before 1000 BC), because there is only one sample in BY2818. However, based on wide STR divergence among Georgian samples in the project, their ancestors may have entered Georgia prior to the recorded history of both the Georgian and Armenian peoples and possibly before their ethnogenesis. If this is correct, then the first I-Y16419 men in the Caucasus were some of the common ancestors of both the Georgian and Armenian peoples.

I-Y16419 ამ მოდგმის უამრავი ადამიანი ამჟამად ცხოვრობს აღმოსავლეთ ანატოლიაში და კავკასიაში, თურქეთში, სომხეთში, ირანში, აზერბაიჯანში, საქართველოში, ჩრდილოეთ ოსეთში, ყაბარდო-ბალყარეთში, ყარაჩაიში. საქართველოშიც: იმერეთში, სამეგრელოში, სვანეთში, ტაოში, ქვემო ქართლში, კახეთში, ხევსურეთში.

ორი გონივრული თეორია არსებობს დაბადების ადგილთან დაკავშირებით I-Y16419 ძვ.წ. 1300.

  1. თუ დაიბადა ბალკანეთში, მისი შვილების ნაწილი წავიდა აღმოსავლეთ ანატოლიაში, სომხეთში, საქართველოში ან ოსეთში ძვ.წ. 1300 წლის შემდეგ. ეს მოსაზრება ეფუძნება იმას, რომ ეს  ხაზები არ არიან დასავლეთ ანატოლიაში, ყველა ემიგრანტი წავიდა ერთად აღმოსავლეთ ანატოლიაში.
  2. თუ დაიბადა აღმოსავლეთ ანატოლიაში ან სომხეთში, შთამომავალი, რომელიც ცხოვრობს ევროპაში, შესაძლოა ესენი არიან საბერძნეთის ან სომხეთის ემიგრანტები.

SNP-ს არსებული მაგალითების ტესტირება ვერ ადგენს საკუთრივ ეთნიკური ქართული შტოების გენეტიკური წარმომავლობის ასაკებს (როგორც ეს დგინდება ეთნიკური სომხებისთვის — ანუ ჩ.წ.-მდე 1000 წ.), რადგან BY2818-ში ქართველის მხოლოდ ერთი ნიმუშია. თუმცა, თუ გავითვალისწინებთ არსებით მრავალფეროვანებას ქართულ ნიმუშების STR-ებს შორის, შეიძლება ვივარაუდოთ რომ მათი წინაპრები საქართველოში უფრო ადრე დასახლდნენ, ვიდრე ამის თაობაზე რაიმე დოკუმენტური წყაროები გაჩნდებოდა (როგორც ქართული ასევე სომხური). და, თუ ეს მოსაზრება სწორია, მაშინ დასაშვებია, რომ როგორც ქართველებში ასევე სომხებში (ზოგადად კავკასიაში) I-Y16419 შტო ერთი კონკრეტული მამაკაცისგან მომდინარეობდეს და ეს მამაკაცი გახდა ქართველებსა და სომხებში ამ მოდგმის (დნმ შტოს) მამამთავარი.

To advance our research into the Georgian I-Y16419 lineages, consider donating to the Georgian DNA project on FTDNA.

Special thanks to Elene Tsutskiridze and Besarion Gugushvili for helping with the Georgian translation.

These posts are the opinion of Hunter Provyn, a haplogroup researcher in J-M241 and J-M102.
Posted in: I2

19 thoughts on “I-Y16419 1300 BC Indigenous or Balkan migrants to Armenia?”

  1. Hello, I am Z-26403+ and presumably BY2818+ (all other downstream SNPs of Z26403 tested -).
    My paternal line is unknown. Autosomally it looks like it is Persian, my most non-maternal matches are Persians from Kuwait and this is what I test on calculators with oracles too.

    I am not sure if this info is of any help. As I see this lineage seems Armenian or Georgian my only theory on how it got to Iran is that it came with large Georgian populations driven there by Shah Abbas in 16. century.

    I will be grateful for any comments or info about more individuals found in Iran of this lineage as I am currently resolving the enigma of who my father was.

    Many thanks

    1. Hi Peter,

      There is a cluster of Iranians in this project all from Karadagh, as you may be aware of, but you do not possess their rare STR values so are not related to them.
      I predict your closest match is Svan Georgian 253677, on the basis of shared rare STRs, but this is not for certain due to high genetic distance at only 37 markers.

      Your theory of descent from the Shah Abbas 200,000 deported Georgians is fascinating. From the wikipedia article – “These deportations marked another stage in the Safavid policy of resettling huge amounts of Georgians and other ethnic Caucasian groups such as the Circassians and Armenians, to mainland Persia. ”

      I don’t know of any other Persians in your lineage but if I find one I’ll let you know.


  2. The date for the most recent common ancestor has again returned to its original 1800 to 2000 BC.
    this was and is stable as it involves more samples in its calculation

    Trialeti culture of the Middle Bronze Age Caucasus seems the most likely origin, part of the larger Kurgan culture.

    Instances of I-Y16419 in the Balkans and Aegean are due to Late MBA migration across the northern half of Anatolia to Southeast Europe, connections with Trialeti culture in Crete and Mycenae include weapons(rapiers, spearheads) and metalwork.
    (source: Militarism and the Indo-Europeanizing of Europe , Robert Drews)

    I-Y16419 is very similar to two subclades of R1b-Z2103, L584 and Y4364, in both TMRCA and distribution, analogous to how I-L801 is connected to R1b-U106 or I-CTS10228 to R1a-CTS1211.

    cases in Germany, Britain, Spain and elsewhere are due to Roman settlements, not necessarily of Italian origin. this is a pattern with many Asian, Balkan, and North African subclades.

  3. Hi. Funny to stumble on this page. I am that BY160929 person from Egypt. I googled my subclade in an effort to find more information about my clade, and ended up here; apparently the only spot on the internet with some information about it. I am stumped as to how my ancestors could have ended up in Egypt, and I’m really keen to learn how and when this could have happened. Is there any way to find out a timeframe? Did we migrate here with the Hittites? Alexander the Great? The Crusaders? The neolithic? The Ottomans? Is there any way to know?

    A few years ago I searched on Ysearch.org and found a person in Saudi Arabia and someone in Iran, both of whom were 1 step away from me on a 25-marker search. The Ftdna Middle East project (مشروع الشرق الأوسط) also has a few people from the Sinai peninsula who have been classified as I-P37. (I was also initially classed as I-P37 until I did further testing). So my amateur non-expert hunch is that there are more of us hiding in Arab countries.

    That’s all I know and wish there was a way to find out more. In the FTDNA “I2b I-L415, I2a2 (was I2c) I- L596” project, I’m classified under my own subcategory. I have no Y-dna matches on FTDNA at any marker level.
    Any advice?

    1. Hi Bou3yan,
      I recommend that you do the YFull analysis. This advances public research because it contains samples tested from all companies. We can then examine how your sample changes the YFull tree and estimated formation and TMRCA dates.

      Then we can look at answering your question.


      1. Hi Hunter,

        I did the Yfull analysis. I’m in the first branch off of I-L596 here: https://www.yfull.com/tree/I-L596/

        As with FTDNA, I have no direct matches. The sibling clades on yfull all appear to be central European, mainly Germany and Italy. There are no hits from Armenia or the Caucasus or Balkans. Though I don’t know if this is just a bias in who gets these tests done…

        There is a new sample in the I-L596 FTDNA Project page who has been categorized with me. He is from Ireland, of all places. Though judging by the STRs, the match is very very distant.

        1. Hello,
          It looks like there is also a new sample from a scientific study on the YFull tree, H26 from Norway. He split I-S6724 just before the end of the bottleneck.
          Interesting about your Irish match. Try using STR Match Finder to see if you share any rare STR alleles with him. Maybe you will find other reliable but distant STR Matches.



          1. Hi Hunter,

            Regarding the Irish person in the FTDNA I-L596 group, he’s been moved into another cluster, so I am alone in there again. Yfull also has me alone and deeply diverged from everyone else, and FTDNA has no y-matches at any level.

            I tried YSEQ’s haplogroup predictor on a whim, and it gives me 66% probability of being J2a-L24 (0.043 score), then J1 and I2. What are your thoughts on that? Is it possible that the I-L596 designation is wrong? I’ve done the Big-Y test on FTDNA, so I think YSEQ’s predictor must be wrong, though it’s interesting that my original result on FTDNA was J1 and it was changed later on.

            Weirdly, when i add more STRs from FTDNA, the results in the YSEQ predictor change. When I go up to 75 STRs, then my top 4 results are J1-FGC3723 (0.07), J1, J2a-L24, and J1-FGC5.

            Regarding the STR matches, I don’t think I’m able to do that without being a customer of YSEQ, and I’m not sure where I should start given my testing history.

          2. I’m sure your SNP test is correct.
            Your STRs are so divergent from the other I2 that the predictor cannot get it right. It’s because your next closest relatives in I2 are 13,800 years distantly related according to the YFull estimate.

          3. Hey, I was actually wrong about the STR match finder. I figured out how to use it. My closest match with all the YSEQ filters selected is a genetic distance of 15 with a 37-STR search. After fiddling with the filters I found that all my hits with a GD of 15 to 17 are in I1, I2, J1, and J2 groups. Interestingly, 2 of them were in Serbian DNA projects and 1 in Slovenian. Don’t know if that means anything at this genetic distance. Several years ago when I was just starting out, I searched on ysearch and found 1 person in Saudi and 1 in Iran who were both GD of 2 away from me on a 25-str search.

          4. If you are the Egypt sample I-Y14158 on YFull then the reason you have no close matches and the closest guys are from different haplogroups is because your next closest relatives are 13,800 years distantly related to you.

  4. It should be noted that Trialeti-Vanadzor culture is believed to have been Indo-European, or at least that it contained Indo-European elements. It disappeared around 1600 BCE, but may have transitioned into Hayasa and/or Metsamor and Lchashen. By the time of its “disappearance” it had spread as far west as Erzerum, Turkey and as far south as Urmia, Iran (or at least their ceramic culture did).

    The spread of the Trialeti-Vanadzor-derived “Transcaucasian” ceramic ware in 1200 BCE corresponded with the attempted invasion of Assyrian by the Urumu and Mushki. They made it as far as Elazig, Turkey, and possibly as far as Cappadocia. There is no presence of western (i.e. European) ceramic ware in eastern Turkey or Armenia at this time.

    The Armenian ethnogenesis completed by around 1200 BCE, according to Nature.

    With these points, and your point about early contact between Armenian and Hurro-Urartian languages that you mentioned in your writeup, seems likely that the Armenian language was present in the South Caucasus by the start of the second millennium BCE.

  5. I am Ashkenazi Jewish I-Z26381, which is part of this group. It exists in many Ashkenazi and Mizrachi Jewish people.

  6. I’m I2c2-L596>Y16649>Y16419, living in Serbia, Belgrade. What do you think about that?

    DYS393-14 DYS390-24 DYS19-15 DYS391-10 DYS385-14-15 DYS439-11 DYS389i-13 DYS392-11 DYS389ii-30 DYS458-17 DYS437-15 DYS448-20 Y-GATA-H4-11 DYS456-15 DYS576-17 DYS570-18 DYS438-10 DYS481-25 DYS549-11 DYS533-12 DYS635-21 DYS643-12

  7. Hi, on what ground do you attribute several Pontic Greeks and Armenians to the branch BY4177? In fact it was defined by my brother’s Big Y test, who is from Bulgaria. He definitely has some close matches, presumed from the same branch, but they are from Serbia, Kosovo, Macedonia and Albania. There are a number of South East Serbians close to the Bulgarian border tested with their own lab, who almost perfectly match by brother’s haplotype. At least one will be tested with FTDNA BIg Y soon, so we will know the age of the branch, or rather the TMRCA, which would not be very old. Don’t see any Pontic Greek too close to him nor any Armenian, the way they are grouped in the Armenian project. Of course this branch has probably come from Anatolia, I2c2 was not found on the Balkans in any old bones until the Roman period. So it was either the Anatolian setters in the Roman cities around the Danubian limes or later with the Byzantine settlers like the Paulicians.

    1. Hi Merab,
      This is quite an old article. I haven’t been keeping track of the development of the research for I-Y16419. Other people who have been tracking it more closely in the last year or so may have better insights. I’ll revisit this later when I have more time to make an update.

Leave a Reply

Your email address will not be published. Required fields are marked *