What percentage of ancestry is enough to make feel part of an ethnic group ?

[Maciamo]

Thanks to population genetics, we are becoming increasingly aware of just how mixed our ancestry really is. The old stereotypes about ethnic purity are meaningless when we think in term of genetic admixtures and deep ancestry spanning over 10,000 years or more.

Yet, ethnic groups still exist and most of us identify with one or several of them. This is because an ethnicity does not require a unique unblended ancestry, but can, and usually is the result of millennia of intermingling in one region. For example, the Germanic ethnicity originated in Scandinavia and North Germany from the admixture of Mesolithic (Ertebølle), Neolithic (Funnelbeaker), Bronze Age (Corded Ware) and Iron Age migrants to the region. As Germanic people later expanded over most of Europe, a lot of Europeans can claim partial Germanic ancestry in addition to other regional ancestry (Celtic, Roman, Greek, Slavic, etc.)

The development of genetic admixtures over the last few years made us realise that most of us have fragments of DNA inherited from ancestors of completely different ethnicities. It is rather common for Europeans to find segments of East Asia, South Asia or East African ancestry in their DNA. And all Europeans have varying shares of Middle Eastern ancestry (which can be subdivided in Anatolian, Caucasian, Mesopotamian, Gedrosian, Southwest Asian, among others).

But just how much ancestry do you feel you need to be able to claim to belong to an ethnic group. This is a question that Americans will surely have thought about more often than Europeans, since Americans (not just from the USA, but the whole Americas) most often have (very) mixed ancestry. That's not a coincidence that ancestry DNA tests were pioneered for the North American market first. Many White Americans, who might pass for "pure Europeans" found out, often to their own surprise that they have a small amount of Native American or African ancestry. Now Europeans can be as surprised to realise that they are not so pure and can also have between 0.1% and 5% of Siberian, or South Asian or African DNA, even if they have red hair, freckles and blue eyes. Looks can be deceptive, as only a small part of our genome has an influence on physical appearance.

We have recently debated the possibility of African immigrants outbreeding Europeans, and that made me wonder about the issue of genetic continuity and the sense of belonging. It is obvious that a child with a European parent and an African one will feel both European and African. One's genetic identity is always clear for the first generation of mixed race children. It gets more complicated after 3, 4 or 5 generations.

Would a child with one European great-grand-parent and 7 African ones feel both ethnically European and African, or just African ? What if it was 1 European ancestor out of 256 (8th generation), amounting theoretically to a mere 0.003% of his/her genome (but perhaps even less) ?

Although all my ancestors as far as my genealogy can go back are European, I tried to imagine what proportion of ancestors of one ethnicity I would need to feel that I can reasonably claim that I partially belong to that group. After some consideration, I thought that 10% was a reasonable amount. That is roughly equivalent to one great-grand-parent (theoretically 12.5%, but we inherit more from some and less from others, so between 10% and 15% from any in average). When I look at my admixtures, I usually feel that 10% is the threshold to which one ethnic/regional category becomes relevant to my identity.

Interestingly, great-grand-parents are also generally the most remote ancestors that we will ever know and remember (very few people have the chance to meet a great-great-grand-parent, and even less to be old enough to remember them). Most people who are not genealogists will also know their grand-parents' names and (hopefully) their great-grand-parents' names, but rarely beyond that. It has also been calculated that the emotional attachment between relatives is only meaningful until the degree of great-grand-parent to great-grand-child, or first cousin (who are the equivalent of a great-grand-parent in genetic distance, also sharing a theoretical 12.5% of DNA with us).

I am interested to know how others feel about this, to see if we all feel more or less the same way, or if there are big interpersonal divergences.

 

[Maciamo]

I noticed an interesting coincidence. 10% happens to be the minimum percentage of Y-DNA haplogroup I1 required for a country or region to be considered (at least partially) Germanic. Have a look at the Y-DNA table yourself. It's uncanny. In France, Flanders-Artois and Alsace have just over 10%, but not any other regions tested. Belgium, West Germany, Switzerland and Austria have all around 12%, and South Germany 10.5%. Scotland has 9%, although the Lowlands are over 12%. Hungary 8%, Wales and Ireland 7%, Brittany 6%, Poland and Lithuania 6%... The only exception is the Czech Republic, which is not considered Germanic but has 11%. Finland is a half-exception has it does have a sizeable Swedish population.

It works less well with R1a and Slavic identity because of the co-existence of Germanic R1a. If we isolated only Slavic R1a, I am pretty sure that 10% would also be an important threshold.

This should also work with J1c3 and Arabic (+Jewish) ethnicity, although the table is for all J1 together. We can see that Turkey, Iran, Armenia and Azerbaijan have all around 9 to 11% of J1, so obligatorily less than 10% of J1c3, while all Arabic countries have over 20% of J1 (except Morocco, which is anyway really much more Berber than Arabic).

10% is also roughly the minimum percentage of European admixture (though not R1a percentage) among Indian Brahmins.

 

[Jackson]

This is quite an interesting subject - I agree with you especially about the 10-15% threshold (or one g-grandparent) being important. I also heard somewhere that oral traditions are often only passed on for around 4 generations or so, which could be connected.

For myself i find that 10-15% works pretty well - Most of my documented ancestry is from England with the remainder from South Wales, the Scottish Borders and Dublin. I identify foremost as English, then British, then Northwest European etc. Or from a different perspective as Celto-Germanic - Although a lot of people like to think they are 100% of the group they identify with a lot of the time, it seems kind of futile when you are dealing with closely related groups like Celtic and Germanic peoples - One thing i've realised over the last few years is that a lot of the time to be Celtic is to also have some Germanic ancestry, and vice versa - So i wouldn't like to say how much of each i am, as there doesn't seem to be a clear definable boundary - or only at the extremities of each group.

It also seems that family rumours of non European ancestry are true - Either from North Africa or the Near/Middle East.
I personally don't feel really connected to that part of my ancestry in the same sense that i do to the rest of my ancestry, probably because of where i've grown up, my language, the history i've grown up with and the group identity me and my family belong to. It also appears to be probably about 6% or so of my ancestry, and so fits in with your observations quite neatly. It's small enough so that if i hadn't known anything about this ancestry i wouldn't have suspected anything, but big enough for me to include it as a significant, if only minor, part of my ancestry even if i don't identify on a personal level with the culture(s) from which it came.

I would say in summary that your idea works well for me - I identify most with my Germanic ancestry, but also with my Celtic ancestry (Hence identifying as Celto-Germanic) - I know about my North African/Near Eastern ancestry and although i can't identify with it as such, i still recognise it as an important part of who I am.


How do you currently consider Belgium in terms of it's ethnic makeup? Although it's almost off topic - i'm curious as you are Belgian yourself, and my father went there often when he was younger.

 

[Maciamo]

I have split the offtopic discussion by ebAmerican about the meaning of ethnicity.

 

[Maciamo]

How do you currently consider Belgium in terms of it's ethnic makeup? Although it's almost off topic - i'm curious as you are Belgian yourself, and my father went there often when he was younger.

Belgians are mixed Italo-Celtic (including Roman) and Germanic, just like the British, North French and many Germans.

EDIT : Actually, as far as Y-DNA haplogroups are concerned, Belgians only have over 10% of I1 and R1b. I1 is obviously Germanic, while R1b is either the Germanic S21/U106 or the Italo-Celtic S116/P312. Within the latter, Belgium has just above 10% of R1b-S28/U152, which could be either Alpine Celtic or Italic/Roman. As Belgian usually claim partial Roman ancestry, and Walloons speak a Romance language, it would make sense if these 10% of S28/U152 were Roman (or perhaps combined with some G2, J2 and E1b1b, as the Romans were themselves a composite). Unfortunately the Y-DNA subclades are not deep enough at present to distinguish Neolithic from Roman G2a, J2 and E1b1b, and ancient Roman autosomal admixtures have not been isolated yet, so it is impossible to say for sure what portion of the R1b-S28/U152 + G2, J2 and E1b1b are Roman anywhere in Europe (even in Italy).

It's also interesting to note that the main difference between Belgium and the Netherlands is that the Dutch have considerably less S28/U152 (max. 6% in the south, and close to 0% in the north).

Any other haplogroup being under 10% of the population, they don't affect the ethnic description of the country.

 

[zanipolo]

Where does this part end in ones ethnicity?

http://en.wikipedia.org/wiki/African_admixture_in_Europe

 

[Maciamo]

Where does this part end in ones ethnicity?

http://en.wikipedia.org/wiki/African_admixture_in_Europe

According to me nowhere because African ancestry is under 10% of any European's genome, and under 10% of any European country or region's gene pool.

This is actually what I wanted to discuss here. I found traces of East African, Southwest Asian, South Asian, Siberian and East Asian DNA in my genome, like about any European would. However that didn't really make me feel entitled to go to Ethiopia or China and tell the locals that I was partly Ethiopian or Chinese. 0.1% (or less) of DNA just isn't enough to feel associated at a personal and meaningful level with an ethnic or racial group.

The question I wanted to raise in this topic was what percentage became significant on a personal level ? I have set the bar to around 10%, and I believe this works just as well for autosomal admixtures in individuals as for Y-haplogroup percentages for a particular ethnic or regional group.

 

[JFWR]

It really depends on what you mean by "10 percent". Do you mean 10 percent in the distant past? 10 percent in the recent past? What?

When I have to succinctly tell people what I am, I say I'm "Celto-Germanic". Do I go around thinking "wow, I'm a Celt and a German by ancestry" all the time? No, because those aren't meaningful ethnicities, but tribal identities of the broad historical foundations of my actual ethnicities.

I'm sure like you, Maciamo, that parts of my DNA, from thousands of years ago, include far flung places. It would be surprising, for instance, if one of my ancestors didn't have a Turkish wife, or had some Egyptian ancestors, or migrated from India, or whatever the case might (but not on my direct paternal line). But I would have no real identification with these identities because it doesn't represent any of my ethnic identities and would be so far in the past that I would have so little connection to it.

 

[Maciamo]

It really depends on what you mean by "10 percent". Do you mean 10 percent in the distant past? 10 percent in the recent past? What?

When I have to succinctly tell people what I am, I say I'm "Celto-Germanic". Do I go around thinking "wow, I'm a Celt and a German by ancestry" all the time? No, because those aren't meaningful ethnicities, but tribal identities of the broad historical foundations of my actual ethnicities.

I'm sure like you, Maciamo, that parts of my DNA, from thousands of years ago, include far flung places. It would be surprising, for instance, if one of my ancestors didn't have a Turkish wife, or had some Egyptian ancestors, or migrated from India, or whatever the case might (but not on my direct paternal line). But I would have no real identification with these identities because it doesn't represent any of my ethnic identities and would be so far in the past that I would have so little connection to it.

I am referring here to autosomal admixtures as tested by the Dodecad Project or Eurogenes. There are many different tools to compare one's DNA with various regions within Europe or around the world. The results are always fairly consistent though. If you haven't tested with 23andMe or a similar company and run your results on the above-mentioned websites, what I am trying to explain won't make much sense though.

Btw, Celtic and Germanic are not ancient or historical ethnicities. They represent actual genetic divisions today as well, because genes don't disappear over time. We call them Celtic and Germanic, because these are the names that best fit the historical territories where those gene pools were originally present, before things got more mixed up.

The main divisions of Europe based on genetic similarities are :

- Celtic (most of Western and Central Europe)
- Germanic
- Balto-Slavic (of which the Uralic speakers can be seen as a subgroup, despite the different linguistic classification).
- Balkan-Carpathian (ex-Yugoslavia, Romania, Moldova and western Ukraine)
- Greco-Anatolian (Albania, Greece, South Italy)

 

[Dorianfinder]

I am referring here to autosomal admixtures as tested by the Dodecad Project or Eurogenes. There are many different tools to compare one's DNA with various regions within Europe or around the world. The results are always fairly consistent though. If you haven't tested with 23andMe or a similar company and run your results on the above-mentioned websites, what I am trying to explain won't make much sense though.

Btw, Celtic and Germanic are not ancient or historical ethnicities. They represent actual genetic divisions today as well, because genes don't disappear over time. We call them Celtic and Germanic, because these are the names that best fit the historical territories where those gene pools were originally present, before things got more mixed up.

The main divisions of Europe based on genetic similarities are :

- Celtic (most of Western and Central Europe)
- Germanic
- Balto-Slavic (of which the Uralic speakers can be seen as a subgroup, despite the different linguistic classification).
- Balkan-Carpathian (ex-Yugoslavia, Romania, Moldova and western Ukraine)
- Greco-Anatolian (Albania, Greece, South Italy)

I have to underscore a problem in your question. For a hypothetical 10% admixture to hold it would have to be possible to distinguish the many ethnic groups from one another in a clear and concise manner. The above groups share much of their admixture and the process is still ongoing.

For example: One will find Germanic and Balto-Slavic admixture very similar in most cases. Greeks and Anatolians will have more Caucasian than the average West European but so do Kurds, Armenians and many Sicilians. This creates more questions concerning gene flow and date of ethnic composition. The labels are not clear enough when dealing with haplogroups, attempting to classify admixture proportions is rather ambitious.

 

[JFWR]

I am referring here to autosomal admixtures as tested by the Dodecad Project or Eurogenes. There are many different tools to compare one's DNA with various regions within Europe or around the world. The results are always fairly consistent though. If you haven't tested with 23andMe or a similar company and run your results on the above-mentioned websites, what I am trying to explain won't make much sense though.

Okay.

Btw, Celtic and Germanic are not ancient or historical ethnicities. They represent actual genetic divisions today as well, because genes don't disappear over time. We call them Celtic and Germanic, because these are the names that best fit the historical territories where those gene pools were originally present, before things got more mixed up.

The main divisions of Europe based on genetic similarities are :

- Celtic (most of Western and Central Europe)
- Germanic
- Balto-Slavic (of which the Uralic speakers can be seen as a subgroup, despite the different linguistic classification).
- Balkan-Carpathian (ex-Yugoslavia, Romania, Moldova and western Ukraine)
- Greco-Anatolian (Albania, Greece, South Italy)

Oh, I am fully aware that Germanic and Celtic genes persist, but in terms of ethnicity these broad designators are useless. Ethnicity is much more recent than the deep ancestry that speaks to the broad strokes of the distant past.

 

[Maciamo]

Oh, I am fully aware that Germanic and Celtic genes persist, but in terms of ethnicity these broad designators are useless. Ethnicity is much more recent than the deep ancestry that speaks to the broad strokes of the distant past.

So what does ethnicity mean for you ?

Personally, I always try to separate genetic ethnicity from modern languages and cultures. That is in part because people can change language and culture, as did non-British immigrants to the USA, although their genes remain the same. Historically, all the Celts lost their language and culture, first to the Romans, then to Germanic people in the British Isles, and a strip of land stretching from Holland to Austria.

In cities like Brussels, Lille or Liège, most of the local populations had become Germanic in early medieval times, but were progressively converted back to Latin and French. In Brussels the language shift only happened in the last 300 years, so that most "native Brusselers" have Flemish names but speak French.

Another example is the triangle formed by the cities of Maastricht, Aachen and Liège, roughly 30km away from one another, and each belonging to a different linguistic and cultural community today, but all part of a same country in the Middle Ages (actually Maastricht and Liège were always part of the same principality or country until as late 1839). Genetically people in these three cities would be undistinguishable from each others.

 

[spongetaro]

I am referring here to autosomal admixtures as tested by the Dodecad Project or Eurogenes. There are many different tools to compare one's DNA with various regions within Europe or around the world. The results are always fairly consistent though. If you haven't tested with 23andMe or a similar company and run your results on the above-mentioned websites, what I am trying to explain won't make much sense though.

Btw, Celtic and Germanic are not ancient or historical ethnicities. They represent actual genetic divisions today as well, because genes don't disappear over time. We call them Celtic and Germanic, because these are the names that best fit the historical territories where those gene pools were originally present, before things got more mixed up.

The main divisions of Europe based on genetic similarities are :

- Celtic (most of Western and Central Europe)
- Germanic
- Balto-Slavic (of which the Uralic speakers can be seen as a subgroup, despite the different linguistic classification).
- Balkan-Carpathian (ex-Yugoslavia, Romania, Moldova and western Ukraine)
- Greco-Anatolian (Albania, Greece, South Italy)

I doubt that "Celtic" is still a relevant genetic division. What people probably mean with it is "Western European" but the genetic similarities between western europeans are the results of Milleniums of shared cultures including Megalithic cultures and the Bell Beakers. The Bronze age Celts only added a few (Gedrosia admixture) to the whole "genetic package" of western Europe.

Before the Celts, there were probably differences in the amount of "Southern Admixture" between Insular and Continental Celts because of stronger Neolithic substratum in the latter. The Cardium Pottery culture and LBK affected much the people of Central and southern Europe. As ancient dna shows, there were still E1b1b in Late antique Bavaria and G2a in early middle ages Bavaria. Even today, there are former historical Celtic lands like Austria and Paris region that kept nearly half of their DNA from the LBK.

We also see that "Southern admixture" is higher in French and Spanish populations than in Irish population for instance.There is no reason to think that it was otherwise during the time of the Celts because Spaniards and Irish people lack Roman and German haplogroups.


The later Germanic and Slavic invasion make Central European Celts (Austria, Bavaria, Czech Republic, Slovakia...), a mix of Slavs, German and Celts with strong Neolithic substratum rather than "Celts".

Then as you said we still don't know wether the R1b U152 found near the Limes of the Roman Empire is Roman or Celtic. People of Eastern France, Southwestern germany and Belgium might be a mix of Roman, German and Celts.

 

[Maciamo]

I doubt that "Celtic" is still a relevant genetic division. What people probably mean with it is "Western European" but the genetic similarities between western europeans are the results of Milleniums of shared cultures including Megalithic cultures and the Bell Beakers. The Bronze age Celts only added a few (Gedrosia admixture) to the whole "genetic package" of western Europe.

I agree that the Proto-Italo-Celtic speakers who arrived during the Bronze Age only represent one genetic component of Western Europeans. However, as they are the last major genetic inflow into the gene pool of Western Europe (before the Germanic migrations, that is), Western Europeans can be considered Celtic. Celtic is not identical to Proto-Italo-Celtic speakers. The Celts are the results of the merger between Megalithic Europeans (already a blend of Palaeolithic Europeans and Neolithic farmers) and Proto-Italo-Celtic speakers from the east (who brought the R1b-S116 lineages). This is what I mean by the term 'Celtic'. I also consider the Italics as a subdivision of the Celts.

However I have to disagree that Proto-Italo-Celtic people "only brought some Gedrosia admixture to Western Europe". I have already explained to you in details that the R1b1b branch that would become associated with the PIE speakers moved north of the Caucasus, where they mixed with the local steppe people (Northeast European admixture, linked to R1a1a). Afterwards, R1b (now a mix of Gedrosia+NE European) moved to Southeast Europe, where they stayed for over 1,000 years, adding a considerable amount of Southeast European + Mediterranean admixtures to their gene pool. After that they settled for a few centuries north of the Alps, where they further mixed with the local population, before conquering Western Europe. So when the Proto-Celtic R1b arrived in Western Europe, they were already heavily hybridised with Palaeolithic Europeans (R1a, I2) AND Neolithic farmers (E1b1b, G2a, and surely also J2 from their home region of Anatolia), so that the Gedrosia admixture might not have exceeded 20-25% of their overall DNA.

The later Germanic and Slavic invasion make Central European Celts (Austria, Bavaria, Czech Republic, Slovakia...), a mix of Slavs, German and Celts with strong Neolithic substratum rather than "Celts".

Then as you said we still don't know wether the R1b U152 found near the Limes of the Roman Empire is Roman or Celtic. People of Eastern France, Southwestern germany and Belgium might be a mix of Roman, German and Celts.

You are agreeing with what I wrote above without even noticing it. It's obvious that modern Europeans are not exclusively Celtic OR Germanic OR Slavic, but that in many places like Austria, Bohemia, Germany of Belgium, people will belong to two or three of these ethnicities. I never claimed otherwise.

The actual point of this thread is to determine how much admixture is the minimum to count as part of one's ethnicity. Some French or Italian people might have a few percent of Slavic admixtures in them, but does that make them truly Celto-Slavic or Italo-Slavic ? I don't think so. If 1% or 2% of DNA is enough to qualify as ethnically part of a group, then all Europeans are Celto-Germano-Slavo-Greco-Sibero-Anatolo-Mesopotamo-Arabo-Africans. That becomes meaningless.

This is why I proposed to set a minimum limit to 10%, and explained why 10% and not 5% or 20%.

With 10% as a limit, almost all Europeans fit in only one or two ethnic categories, and a few (notably in the former Austro-Hungarian Empire) in perhaps three ethnicities.

 

[JFWR]

So what does ethnicity mean for you ?

Personally, I always try to separate genetic ethnicity from modern languages and cultures. That is in part because people can change language and culture, as did non-British immigrants to the USA, although their genes remain the same. Historically, all the Celts lost their language and culture, first to the Romans, then to Germanic people in the British Isles, and a strip of land stretching from Holland to Austria.

In cities like Brussels, Lille or Liège, most of the local populations had become Germanic in early medieval times, but were progressively converted back to Latin and French. In Brussels the language shift only happened in the last 300 years, so that most "native Brusselers" have Flemish names but speak French.

Another example is the triangle formed by the cities of Maastricht, Aachen and Liège, roughly 30km away from one another, and each belonging to a different linguistic and cultural community today, but all part of a same country in the Middle Ages (actually Maastricht and Liège were always part of the same principality or country until as late 1839). Genetically people in these three cities would be undistinguishable from each others.

My definition of ethnicity would be this, broadly speaking:

Your national extraction of your family going back 300-500 years, provided that they were part of that nation (i.e. not Jews, Gypsies, Druze, or other perennial outsiders) for that duration. Specifically, your ethnicity might also include a special designation because of a further historical migration or identities, like say...Hugenot.

For areas that were previously unique countries, it would be most proper to speak of a local identity as your ethnicity (provided your family comes from there and it is not a movement after national unification). So, for instance, one would be properly Bavarian, or Venetian, or Flemish rather than German, Italian, or Belgian.

For Americans, excluding those who came here with the Mayflower and other early immigrants, almost no one is ethnically American. They are nationally American, but ethnically Irish, English, French, Jewish, Chinese, West African or whatever the case might be.

I think the term ancestry would accord well with Celtic, Germanic, Turkic, Meso-American, or whatever the case might be. These represent your deep roots that do not represent ancient ties.

 

[spongetaro]

I agree that the Proto-Italo-Celtic speakers who arrived during the Bronze Age only represent one genetic component of Western Europeans. However, as they are the last major genetic inflow into the gene pool of Western Europe (before the Germanic migrations, that is), Western Europeans can be considered Celtic. Celtic is not identical to Proto-Italo-Celtic speakers. The Celts are the results of the merger between Megalithic Europeans (already a blend of Palaeolithic Europeans and Neolithic farmers) and Proto-Italo-Celtic speakers from the east (who brought the R1b-S116 lineages). This is what I mean by the term 'Celtic'. I also consider the Italics as a subdivision of the Celts.

However I have to disagree that Proto-Italo-Celtic people "only brought some Gedrosia admixture to Western Europe". I have already explained to you in details that the R1b1b branch that would become associated with the PIE speakers moved north of the Caucasus, where they mixed with the local steppe people (Northeast European admixture, linked to R1a1a). Afterwards, R1b (now a mix of Gedrosia+NE European) moved to Southeast Europe, where they stayed for over 1,000 years, adding a considerable amount of Southeast European + Mediterranean admixtures to their gene pool. After that they settled for a few centuries north of the Alps, where they further mixed with the local population, before conquering Western Europe. So when the Proto-Celtic R1b arrived in Western Europe, they were already heavily hybridised with Palaeolithic Europeans (R1a, I2) AND Neolithic farmers (E1b1b, G2a, and surely also J2 from their home region of Anatolia), so that the Gedrosia admixture might not have exceeded 20-25% of their overall DNA.



You are agreeing with what I wrote above without even noticing it. It's obvious that modern Europeans are not exclusively Celtic OR Germanic OR Slavic, but that in many places like Austria, Bohemia, Germany of Belgium, people will belong to two or three of these ethnicities. I never claimed otherwise.

The actual point of this thread is to determine how much admixture is the minimum to count as part of one's ethnicity. Some French or Italian people might have a few percent of Slavic admixtures in them, but does that make them truly Celto-Slavic or Italo-Slavic ? I don't think so. If 1% or 2% of DNA is enough to qualify as ethnically part of a group, then all Europeans are Celto-Germano-Slavo-Greco-Sibero-Anatolo-Mesopotamo-Arabo-Africans. That becomes meaningless.

This is why I proposed to set a minimum limit to 10%, and explained why 10% and not 5% or 20%.

With 10% as a limit, almost all Europeans fit in only one or two ethnic categories, and a few (notably in the former Austro-Hungarian Empire) in perhaps three ethnicities.

I didn't disagree with your minimum admixture part, I'm just arguing about.

 

[Maciamo]

My definition of ethnicity would be this, broadly speaking:

Your national extraction of your family going back 300-500 years, provided that they were part of that nation (i.e. not Jews, Gypsies, Druze, or other perennial outsiders) for that duration. Specifically, your ethnicity might also include a special designation because of a further historical migration or identities, like say...Hugenot.

For areas that were previously unique countries, it would be most proper to speak of a local identity as your ethnicity (provided your family comes from there and it is not a movement after national unification). So, for instance, one would be properly Bavarian, or Venetian, or Flemish rather than German, Italian, or Belgian.

It is hard for me to connect ethnicity to political countries because :

1) political borders are artificial creations, usually resulting from war of conquests and inheritance through political marriages. For example, France expanded it borders by conquering regions that once belong to the Low Countries (Flanders-Artois), to Germany (Lorraine, Alsace, Franche-Comté), or to Italy (Corsica), and acquired others that were not culturally or linguistically French either through politcial arrangements (Brittany, Savoie, Nice). So why should a Breton, an Alsatian, or a Corsica feel ethnically French just because all their ancestors were French citizens in the last two or three centuries.

2) political borders often don't match geographic divisions of the genetic ancestry. This is especially true in big countries like Russia, Iran, India, China or Indonesia, where there are plenty of completely different ethnicities, yet all belong to a single nation or country, and in some cases (Iran, India, most of China) have belong to that political entity for several millennia.

3) political borders change over time. Belgium was only created in 1830. Before it has belong to Rome, Germany, France, Burgundy, Spain, Austria, the Netherlands... Parts of Belgium were German while others were simultaneously French or Spanish. Some speak (or spoke) Flemish Dutch, Brabantine Dutch, Limburgs Dutch, Walloon, Picard, French, and Franconian German. Yet people have always felt part of the Southern Netherlands (along with Luxembourg and the Nord-Pas-de-Calais region of France), part of a common ethnic and cultural group despite differences in nationalities and languages. The cuisine, architecture, lifestyle, customs, and mindset have remained suprisingly similar in all the historical Southern Netherlands despite political divisions that subsist to this day. The Nord-Pas-de-Calais was annexed to France 350 years ago, but still feels much much more Belgian in every respect than French. 350 years wasn't enough to change the local culture. It also didn't change the gene pool much. More people from Nord-Pas-de-Calais migrate to Paris than the other way round.


That is basically why I cannot link ethnicity to political countries.

 

[Maciamo]

I didn't disagree with your minimum admixture part, I'm just arguing about.

Do you agree with 10% (approximately) as a minimum ? If not what do you suggest ?

 

[spongetaro]

Do you agree with 10% (approximately) as a minimum ? If not what do you suggest ?

10% seems to work pretty well with I1 (though we could have add U106 too) and the German "ethnicity". However, the dominance of one haplogroup in a certain area (typically R1b P312 in western Europe) is quite misleading as several ethnicities can "hide" behind it. You maybe get a more accurate picture by using the 10% treshold and the autosomal admixture. It seems to work well with the Gedrosia admixture notably. Autosomal admixture can thus reveal the different "ethnicities" that hide behind R1b that is to say: the different stratas that shaped the western European genetic package over millenia.

 

[JFWR]

It is hard for me to connect ethnicity to political countries because :

1) political borders are artificial creations, usually resulting from war of conquests and inheritance through political marriages. For example, France expanded it borders by conquering regions that once belong to the Low Countries (Flanders-Artois), to Germany (Lorraine, Alsace, Franche-Comté), or to Italy (Corsica), and acquired others that were not culturally or linguistically French either through politcial arrangements (Brittany, Savoie, Nice). So why should a Breton, an Alsatian, or a Corsica feel ethnically French just because all their ancestors were French citizens in the last two or three centuries.

Excluding the Germanic areas of modern France, I would say there is only a small distinctiveness for these communities that has been part of another nation for some 300 years. Bretons, for instance, are not especially "non-French" because they have been French for centuries. One can be ancestrally Breton, but it is highly unlikely you can be ethnically Breton in any meaningful sense. Just as, for instance, Calais is not ethnically English anymore.

2) political borders often don't match geographic divisions of the genetic ancestry. This is especially true in big countries like Russia, Iran, India, China or Indonesia, where there are plenty of completely different ethnicities, yet all belong to a single nation or country, and in some cases (Iran, India, most of China) have belong to that political entity for several millennia.

Indeed, which is why I suggested one could be an __________-___________. So one might be a Tartar-Russian or a Canton-Chinese.

3) political borders change over time. Belgium was only created in 1830. Before it has belong to Rome, Germany, France, Burgundy, Spain, Austria, the Netherlands... Parts of Belgium were German while others were simultaneously French or Spanish. Some speak (or spoke) Flemish Dutch, Brabantine Dutch, Limburgs Dutch, Walloon, Picard, French, and Franconian German. Yet people have always felt part of the Southern Netherlands (along with Luxembourg and the Nord-Pas-de-Calais region of France), part of a common ethnic and cultural group despite differences in nationalities and languages. The cuisine, architecture, lifestyle, customs, and mindset have remained suprisingly similar in all the historical Southern Netherlands despite political divisions that subsist to this day. The Nord-Pas-de-Calais was annexed to France 350 years ago, but still feels much much more Belgian in every respect than French. 350 years wasn't enough to change the local culture. It also didn't change the gene pool much. More people from Nord-Pas-de-Calais migrate to Paris than the other way round.

Thus why I said 300-500. Some areas are not especially good at becoming distinctly part of another ethnicity.