November 01, 2011

Roman DNA Project

Kristina "Bone Girl" Killgrove is launching a Roman DNA Project
This is a pilot project to study the DNA from the skeletal remains of lower-class Romans dating to the Imperial period. Some DNA analysis has been done in ancient Italy, but it hasn't yet been explored as a method for understanding the geographic origins of immigrants to Rome. In particular, the combination of DNA with results I obtained from isotope analyses (Sr/O/C/N/Pb) in my dissertation research will be a powerful way of investigating the life histories of groups under-represented in history, particularly slaves, immigrants, women, and children.
She is asking for contributions to the Roman DNA Project. There is also a video associated with the Project:

This is a good idea, and I encourage readers to contribute, if, after reading the project materials, they want to pitch in.

Just one quick comment: testing lower-class Imperial Romans is, indeed, admirable, but I am pretty sure there are ancient remains from Latium that date to before the period when cremation burial became the norm, with inhumation never entirely dying out even among the patrician gentes.

In studying the origin of foreigners in Rome, it makes sense to first have a picture of what the Romans of old were like. They may not have been "under-represented" in history, but they are certainly under-represented when it comes to ancient DNA research.


Kristina Killgrove said...

Oh wow, thanks so much for the shout-out! Your comment about older burials is well taken. I actually have samples from three different periods, ranging from the 4th century BC to the 3rd century AD. The Imperial sample is the largest, so that's what I've been focusing my efforts on. But I may see if I have enough funds to test some of the earlier people too! Of course, the more people donate, the more testing I can do. :)

Thom said...

We will get a comment questions of history.

Dienekes said...

Kristina, what type of DNA testing do you plan to do? Many people who visit my blog are very interested in mtDNA and Y-chromosome testing, rather than the more disease-side of things, and I am sure many of them would be willing to finance testing of ancient Roman remains from the Republic and Imperial eras.

It's a shame that we now have at least 4 different Neolithic samples of Y-chromosomes from Europe, and apparently none from the classical Greeks and Romans. Any effort in that direction is commendable.

mooreisbetter said...

This lacks any precision. What was a "Roman" was defined five different ways over the course of Rome's existence.

1. There were the original Latin Romans, which felt themselves different from their closest neighbors, the Etruscans and Sabines. This would be the most precise definition of an original Roman and Roman DNA.

2. There were the pre-Imperial Roman citizens, a burgeoning population of Latins with soe Etruscan and Oscan admixture, who existed for hundreds of years and populated many colonies, before mass immigration TO Rome.

3. There was a period when all Italians were defined as Roman citizens.

4. After the reign of Caracalla, "Roman" citizenship was extended to all Romans within the boundaries of the Empire.

5. During Byzantine times, the definition got cloudier still, with the term becoming somewhat more attenuated.

After reading this, it appears that she is trying to identify ying to identify the DNA of "Romans" in #4.

That is inherently imprecise, and the project is misnamed. It's not a study of Roman DNA -- it's a study of what DNA came to the Roman Empire. I'd analogize Britain or America, but most people are smart enough to know this.

I suggest working with a historian to identify the DNA composition of Romans in, for example, the Republican and early Imperial Times, when a "Roman" was defined as an Italian of Latin or mixed Oscan ancestry residing in Italy or one of the Roman colonies?

You don't even need Ancient DNA to do it. Just a thorough grasp of history.

There are towns that exist in Italy, France, Spain, and other places that were settled by Romans before the Empire.

That is, these were the "original" Latin/Italian Romans, before there were many immigrants to the area. If you make a list of those colonies, and then run that list through a grading system (offering points for remoteness, subtracting points for subsequent invasions), you will come up with certain modern populations that are Roman-descended.

For example, if you take (1) an isolated town in Latium, settled by Romans, with no history of slave-based agriculture, and then (2) a very isolated mountain village of no importance, settled by Romans and not ever invaded by Goths, Saracens, Normans, etc., and (3) a similar town in, say, Spain -- any genetic similarities between the three would be more than coincidence, and would be the closest we could come to approximating the Ancient Roman genome.

This is the study I'd like to see.

Kristina Killgrove said...

@Dienekes - I plan to do mtDNA analysis, since I want to look into the haplotypes/haplogroups for the people I think arrived at Rome during the Imperial period as immigrants. In my previous research, I've done Sr/O/C/N/Pb isotope analysis and am fairly confident that 20 individuals (out of the 55 I could afford to subject to all those isotopes) came to Rome as immigrants (free or enslaved).

I'm interested in Y-chromosome testing, but as I understand it, it's expensive. As is DNA analysis for disease (which eventually I hope to explore, since the question of malaria and various genetic anemias is quite interesting). Since this is very much a pilot project, I may have to reserve those kinds of expensive/painstaking DNA analyses for the near future.

Dienekes said...

1. There were the original Latin Romans, which felt themselves different from their closest neighbors, the Etruscans and Sabines. This would be the most precise definition of an original Roman and Roman DNA.

The Romans did not feel themselves different from the Sabines. The Sabines were reckoned as part of the Romans' ancestry.

Also, we should be careful when speaking about "Latin Romans" for the early age prior to the founding of Rome, conflating the ethnic group with the language.

The Romans themselves held the belief that some of their ancestors had come from Troy and had intermarried with the Latins, an event remembered as the marriage of Aeneas with Lavinia, the daughter of King Latinus.

I often parallel the case of the Romans with that of Bulgarians. We have Bulgars (Turkic speakers) + Slavs + pre-Slavic Thracian inhabitants that gradually formed a nation that got its name from one source, its language from another, and a lot of its DNA from a third source.

Kristina Killgrove said...

Thanks for your comment, Christina. You're right that I'm not being at all precise about the term "Roman" because I'm putting this out there to the public at large rather than applying through a granting organization that will have experts read a proposal.

My project is not about identifying who would have considered themselves "Roman," though. You are, of course, right to point out that that status changed through time and primarily related to citizenship. My project is about doing mtDNA analysis of people who were buried in Rome between the 1st-3rd centuries AD.

I have already done isotope analysis of remains from 100+ people buried at Rome in the Imperial period. Roughly 1/3 of them had isotope signatures that placed their origins outside the larger Roman suburbs. I unfortunately cannot pinpoint those origins with isotopes, but mtDNA may help. That said, depending on what the results show, I may only be able to say "all these people were from south-central Europe." Still, since several of the isotopes suggested origins in northern Europe and Africa, I am hopeful that the DNA analysis will reveal the geographic origins of people who died at Rome. This information is not available through history - I use the historical record quite a bit to form hypotheses, especially about structural factors that would have affected immigration and slavery.

As Dienekes suggested, there hasn't been much DNA work done in this area. It's been prohibitively expensive until recently, and many people don't realize that there are thousands of skeletons from the Imperial period in labs around Rome.

If you're interested, you can check out my dissertation, which engages with the historical and epigraphical literature far more thoroughly than I could do in a 3-minute video aimed at non-specialists: K. Killgrove. 2010. Migration and mobility in Imperial Rome. UNC Chapel Hill. (It's available as a PDF from my website at

Dienekes said...

I think the question of whether to label the tested individuals "Roman" or not is beside the point. The important thing is data: where/how was someone buried, grave goods, isotope analysis and DNA.

By combining different sources of evidence, a clearer picture of the origins of individuals will emerge.

A reason why I think it is important to test pre-Imperial Romans is that ancient DNA testing has increasingly presented us with a picture of substantial changes in haplogroup frequencies in many parts of the world. So, if we want to discover how people moved, data on modern populations may not always be a safe bet.

For example, ancient Etruscan mtDNA shows parallels to West Asia and differences from modern Tuscans. If we did not have that mtDNA, we might interpret a West Asian-like mtDNA in Rome as a recent immigrant from Anatolia, when, in fact, he could be of Etruscan origin. Or, recently there were three studies showing "East Asian" DNA in Neolithic Hungary, Ukraine and Imperial Age Rome. We may have interpreted the Imperial Age Rome mtDNA as that of an exotic immigrant, but that interpretation changes a lot if such mtDNA was present in Europe since the Neolithic.

But, the bottom line is: data is good, and any effort to get more data is good.

Kristina Killgrove said...

@Dienekes - For the moment, the question of whether someone was "Roman" is beside the point, but since I'm an archaeologist, it would be great to explore the questions of ethnicity and identity in the future. After I get some more data.

It looks like I have a whole lot more reading ahead of me! I'm really excited about this project, and I hope it gets funded so that I can undertake it.

mooreisbetter said...

I hate to beat a dead horse, but Kristina, words matter and terms have a meaning. This is being pitched as a "Roman" DNA project when really, it is nothing of the sort.

(See discussion below on what is a "Taino" if you want to see just how anal we can be on this).

@Dienekes: I have been very respectful of those who know more than me on this board, but know you are talking to a specialist in this subject.

The early Romans did NOT consider the Sabines as "Roman." Numerous late Republican sources note how the Roman nobility, which was mostly descended from Latins, gave the Claudii grief for being descended from Sabines. Family patriarch Attus Clausus hailed from Sabine territory. Ditto for the Porci Catones (and their most famous member, Cato the Censor).

And the foundation mythology (claiming ancestor from Trojans and Greeks) was just an attempt by a much younger, less accomplished people to tie themselves to a very well-established, learned people: yours, Dienekes -- the Greeks.

Dienekes said...

The early Romans did NOT consider the Sabines as "Roman." Numerous late Republican sources note how the Roman nobility, which was mostly descended from Latins, gave the Claudii grief for being descended from Sabines.

Are late Republican sources evidence about what early Romans thought?

The Romans were aware of the mixed origins of their city and its citizens. Of course they identified most strongly with the Aeneas-Romulus line of descent, but they were aware that these Proto-Romans, so to speak, had incorporated different elements in the city, the Aborigines of King Latinus, and the Sabines being the most important, while Etruscans were also important but were probably downplayed by the Romans of historical times as their Republic was born and flourished in opposition to the Etruscans.

idurar said...

Well, I don't care if they are native or not. They were buried in Rome, that's the most important and the only thing I care about. It already would tell us a lot about Ancient Rome.
When we'll have the data, we can discuss about the origins of the haplogroups.

So, if I am not wrong, only mtdna will be tested. Of course, I hope the y-dna too will be tested, someday.

Kristina Killgrove said...

@Christina - The term "Roman" definitely has meaning, but that meaning is different depending on time period. You can correct me if I'm wrong, but histories primarily give us evidence of the legal use of the term "Roman." Much less understood is the social construct of Roman, since this would have varied in both time and place and since we don't have written records about the vast majority of the people in the Empire. Did some guy in Dacia or Greece or Asia Minor consider himself "Roman"? Was it an insult hurled at him? Did it help him rise in status? Some of this can be worked out through epigraphy and history, but those two lines of evidence still leave vast holes - holes that Roman archaeologists (or can I not call myself that?) like me are trying to fill.

Isotope work is giving me a good idea of the geographical origins of people at Rome in the Imperial period, particularly in combination with history, epigraphy, and burial practices, but DNA analysis is the next step.

What's interesting to me is that foreigners at Rome are doing the same things as locals: eating the same food, catching the same diseases, burying one another in the same way. It'll be especially interesting to look at the DNA of both groups - those that I think are locals (and therefore either properly Romans or slaves for a Roman family) and those that are foreigners. Tracy Prowse's study showed that there was an Asian in rural southern Italy. Imagine what the DNA will show about urban Rome - it may very well show us that people came there from areas far outside what we consider to be the boundaries of the Empire.

So that's the point of the Roman DNA Project - to study the biology of people who were buried in Rome in the general "Roman" period.

mooreisbetter said...

My god Dienekes, you can be so argumentative. Yes, if the Romans living in 50 BCE were aware of differences between Latins and Sabines, and the Romans in 100 CE were mindful of the same, it follows a fortiorari that Romans in 300 BCE were aware of these same differences.

Plus it's a fact. Read your Polybius and your Livy and your Suetonius and then we'll talk again.

Regardless, your odd hair-splitting here doesnt diminish the poinrt that a "roman" in early Republican times was generally thought of as a citizen of Rome of ancestry in the founding tribes, all central Italians.

In late Republican and early Imperial times, this definition encompassed Italians and some descendant of colonists in modern France and Spain.

In later Imperial times, the definition encompassed many others, who lived in the boundaries of the empire.

This is not unlike the evolving definition of what a true Greek was. The Ionians considered themselves the purest and original. The Dorians came next, and were accepted into the definition. Macedonians next, but many did not consider them Greek enough. By Roman times, Hellenized Syrians were called "Greek."

If you can define Taino buddy, trust me, I can define Roman at any stage in time from 753 BCE to 1453 CE.

Dienekes said...

My god Dienekes, you can be so argumentative. Yes, if the Romans living in 50 BCE were aware of differences between Latins and Sabines, and the Romans in 100 CE were mindful of the same, it follows a fortiorari that Romans in 300 BCE were aware of these same differences.

Are you defining "early Romans" as the Romans who lived in 300BC?

And, no, I don't see how it follows. What we know about Roman history is that the Roman identity solidified out of a people of diverse origin, and then expanded as the Romans started expanding.

In any case, acknowledging the mixed origins of Romans, in accordance with what they actually said themselves is not incompatible with their having developed a disliking for non-Roman elements in their ancestry later on. A good example is that of the Turks, who are partly of Greek origin, named their sultanate "Rum", but today mostly view themselves antithetically to Greeks/Westerners and emphasize either their native Anatolian or Central Asian origins.

Anonymous said...

We must be careful: the Romans knew who they were and what the Sabines was the difference in language and customs with them, but that does not mean that there was a large genetic difference with them.
I bet that most of the inhabitants of imperial Rome were Italics, even though we know that there were foreigners, like the Jews.
The people ate and bought the same things, because the Roman people lived on government subsidies, so many immigrants came from the countryside, when the estate began to employ slaves.

mooreisbetter said...

You're way off. And nothing you just said detracts from my original point: that the definition of Roman changed over time.

Roman identity suffered from the opposite of your example. Everyone wanted to be a Roman< then and now.

Kristina Killgrove said...

@DagoRed - I think you're right that most of the people living in Imperial Rome were from the peninsula - that's what the historical record suggests as well as what my (admittedly small) sample of skeletons revealed through Sr/O isotope analysis. At different times, there may have been diasporas of people to Rome (as slaves), such as after any major war.

Interestingly, although the people living in Rome ate mostly the same diets (based on wheat, terrestrial foods, and terrestrial animals), the C/N isotope data show surprising variation even in the Roman suburbium. And I found that many immigrants to Rome ate a diet high in millet in childhood - perhaps the "millet and beans" diet that Pliny talks about the rural Italic folk eating. So the government dole of wheat (and, later in time, pork) likely influenced the diet of the lower classes, but the way they chose to supplement their diets shows variation that I think is quite interesting.

mooreisbetter said...

I agree Dago Red. Thanks for answering our questions Kristina, and thanks for a spirited exhange Dienekes.

alfio said...

Mixed origin?
Here we are talking of central Italian tribes with probably slight different traditions, not of Chinese from one side and Swedish from other. Melting would be a better term than "mixing" when we talk about tribes as Sabines and Latins.