In the early days of the empire, Rome was big, probably around 1 million. There were some number of Jews (thousands at least, perhaps as many as 40,000 by some estimates, although that’s probably high). There were also other foreigners in Rome – probably mostly Greeks, but also some Gauls and Syrians and such. There were also many slaves in Italy, perhaps a third of the population.
Mostly those slaves were from Europe, obtained in wars of expansion: Gaul, Hispania, Germany, Britannia, Greece, etc. Wiki suggests that their European origins is why they didn’t have much affect on Italian genetics because they were European, but that’s wrong. You can certainly detect genetic differences between Gauls and Italian, Germans and Italians, etc. Wiki is correct in saying that you don’t see much sign of this ancient immigration in Italian genetics, but it’s not because they were just like Italians: it has to be because they died out.
The cities were population sinks, and collapsed with the Empire. I doubt if slaves had high birth rates: certainly those working in mines or quarries didn’t. Nor did gladiators. One way or another, the foreigners in Italy, the vast majority of them, disappeared.
Now you see some signs of other stuff in Sicily or Calabria today, but that seems to be later, from Arab or Byzantine times.
West Hunter 
Sounds good. Unfortunately, ancient DNA says otherwise, so I suggest you change the title of your post to “No longer Italian”:
http://www.ncbi.nlm.nih.gov/pubmed/19570745/ – “Genealogical discontinuities among Etruscan, Medieval, and contemporary Tuscans”; Tuscan- Etruscan question aside, in a particularly relevant to the post section, research suggests discontinuity between the population of the Roman Empire and contemporary Italians and continuity between early medieval (post barbarian invasion) Italians and contemporary Italians.
Similar relation is argued here as well – http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4312042/ – “Genealogical Relationships between Early Medieval and Modern Inhabitants of Piedmont” – but more cautiously, however, authors detect mt dna connection between longobard population of early med. epoch and contemporary Piedmont; but the first research is much more clear – population replacement in Toscana after barbarian invasions. So I suggest the owner of this blog reconsider his views on the population history of Mediterranean region from antiquity till’ now very, very seriously.
MtDNA only tells half the story, and not very well. Etruscans weren’t the only inhabitants of ancient Tuscany. They were a minority conquered by the Italic Romans, so it’s not surprising that later Tuscans don’t match them exactly. It has nothing to do with barbarian invasions, which better methods show didn’t have much of an impact on Italy.
Low coverage mtDNA can rarelly tell more than the large geographic region someone is from; like Europe or maybe just West Eurasia. 27 mtDNAs from Etruscans tell us just about nothing about how they’re related to modern Tuscans.
um, they sequenced 11 mtDNA samples spanning from 10th century BC – 1th century AD, which includes republican period of Rome and placed the results against the mtDNA of 27 extractions dating 10th century AD-15th Century AD and then they placed them against a pool of 347 people from contemporary toskania; results – continuity of mtdna between 1000 bc – 0 ad; then italian mtdna in toskana disappears and is substituted by the mtdna found in 10th century ad graves. Continuity b/y mt dna from 10th century toskana and 21th century toskana. Results encompass not only etruscan period, they encompass roman republican period as well. Population replacement somewhere between 1century AD-10th Century AD. Clearly. Until better ancient dna research refutes it, the results hold, unless you know a better method of deciphering the population history of a region than ancient dna sequencing. I would be curious to learn what that method might be.
There is an autochthonous bias in every human being. Being autochthonous is an instinctive claim for legitimacy – so it is attractive. Everyone has it. But, as the authors put it, “Extensive demographic changes before AD 1000 are thus the simplest explanation for the differences between the contemporary and the Bronze-Age mtDNAs of Tuscany. Accordingly, genealogical continuity between ancient and modern populations of the same area does not seem a safe general assumption, but rather a hypothesis that, when possible, should be tested using ancient DNA analysis.
PMID: 19570745 DOI: 10.1093/molbev/msp126″. Not much to add.
I quoted from a study that uses a better method than mtDNA sequencing. There was no population replacement by barbarians in Tuscany or anywhere else in Italy.
Modern Tuscans are genetically closer to the few ancient Etruscan genomes we have than they are to any Germanic people.
The population of Rome could not have ever been higher than 1/2 million before the modern era -not enough area. Most likely 1/3 of a million. First city over 1 million was probably Chang’an.
The Aventine Wall enclosed 436 hectares, small compared to a modern city.
95% of Romans lived in insulae, multistory tenements.
Insulae were sometimes up to 6-7 stories high. Base area was around 3200-4300 square feet.
In the late fourth century, there were 46k insula in the city – probably more than that in 100 AD.
You could have easily crammed a million people into Rome: that fits the known grain imports.
Do I take it that the figure of 1 million is someone’s modern estimate rather than the result of a Roman census?
If the “40,000” figure is just one of those figures tossed around from ancient times I’d take it as an overestimate, like so many other ancient figures for the sizes of armies, cities, and so on.
“So how many Jews were there in the Roman Empire? What fraction were they of the population? Nobody knows. We have possibly-ok estimates for Rome at certain points in time. Two comments by Josephus suggest 4-8 thousand – if you believe him – and one by Tacitus mentioned 4,000. That may mean adult males only: 8 thousand adult males would imply a total population of something like 40,000 in Rome. Or maybe not. There were certainly Jews in classical Rome, enough to sustain a number of synagogues, including a Samaritan synagogue.”
From the Regionary Catalogue, an existing document from the late 4th century,
46,602 insulae and 1797 buildings identified as a domus (single-family home).
What city in the world today has an even remotely similar population density?
Parts of Hong Kong or Calcutta, maybe.
Colin McEvedy was a historian with a strong interest in population, here is what he had to say on the subject:
“Calculations of ancient Rome’s population depend on four bits of information. The first is the result of the city’s housing taken in Constantine’s day. It gives the number of houses as 1790, and the number of apartments as 46,000. On the basis of ten persons to a house and four to an apartment, this yields […] 201,900.
The second is the number of people who got a free wheat ration from the state. When the rosters were monitored properly, as they were by Julius Caesar and Augustus, they totalled 150,000 and “a few more than 200,000″ respectively.
The third is the census of AD 1526, when 45,178 people lived in … 220 hectares … Applied to the walled area of ancient Rome (1380 ha) this yields a population figure of 283,000. However, something on the order of a third of the classical city was public space of one sort or another (gardens, temples, baths) which brings us back to 200,000 again.
Finally, there is the size of the city when it did reach the million mark in AD 1931. At 6780 ha. it was bigger than ancient Rome by a factor of 4.9, suggesting a population of 204,000 for the classical city.
The million men deal with this data in some pretty contorted ways. First, they say that apartment (insula) is really an apartment block, which enables them to use any multiplier they fancy. This is not allowable. The minimum ground floor space of an apartment block is of the order of 300m^3 (0.03 ha); 46000 of them would occupy the entire area within the Aurelian walls …
As to the 200,000 or so who got the wheat ration, the million men say that they were adult males and need a multiplier of 4, plus a bit more for slaves. This won’t do either; the only named recipient we know of is a woman, and reliefs on the Arch of Constantine show that the queues for imperial handouts included women and children …
So 200,000 it is, as Augustus said, and indeed who ever heard of a dictator who put a smaller figure on his largesse than he needed to. If he had fed a million Romans he would have said so.”
[From “The New Penguin Atlas of Ancient History”, 2002]
I will also note that the second largest city in the Mediterranean world was Alexandria with 70,000-90,000 inhabitants. It would be strange, though not of course impossible, for Rome to be 11-14 times larger than the second city of the Empire (itself an extremely large metropolis and perhaps the largest city the world had ever seen in its earlier Ptolemaic heyday).
Thanks for that: fascinating. I’ve got three of McEvedy’s excellent books. It was only when I saw his obit in the Telegraph that I learned that he was a psychiatrist by trade.
I have a number of McEvedy’s books: his stuff is a blast. Was he right about his low-ball population estimates, in particular for Rome?
I don’t know. I dug into it in response to your comment, and now know way more the Regionaries than I ever wanted to: but I can’t tell. I doubt if McEvedy was right in thinking that “insula” meant an apartment rather than an apartment building, but it is clear that the Regionaries exaggerated the size of some checkable things in Rome.
“The million men deal with this data in some pretty contorted ways. First, they say that apartment (insula) is really an apartment block, which enables them to use any multiplier they fancy. This is not allowable. The minimum ground floor space of an apartment block is of the order of 300m^3 (0.03 ha); 46000 of them would occupy the entire area within the Aurelian walls …”
My understanding is that the insulae were apartment buildings, not apartment blocks. Hence, the major multiplier would involve the vertical, not the horizontal. And since many of the insulae reached 6-7 stories in height, that would be a significant factor indeed.
The population density of Naha, Okinawa is 8244/sq. km.
A million people in 400 hectares would be about twice the population density of Manilla.
I picked up the number from the wrong column in Wikipedia. Actually it would be about 5 times the population density of Manilla.
Better size would be within the Wall of Aurelian: 1400 hectares.
While I don’t really know much about it I would guess that the population was probably well under a million.
That squares up with the evidence I’ve seen.
Few Foreigners in Ancient Rome
http://italianthro.blogspot.com/2010/09/few-foreigners-in-ancient-rome.html
Minimal Demographic Impact of Roman Slavery
http://italianthro.blogspot.com/2010/09/minimal-impact-of-roman-slavery.html
Moors Expelled from Sicily and Southern Italy
http://italianthro.blogspot.com/2011/10/moors-expelled-from-sicily-and-south.html
Affinity of Ancient and Modern Italians
http://italianthro.blogspot.com/2012/07/affinity-of-ancient-and-modern-italians.html
I could see it being that as effective urbanisation declines in Italy, few people actually move out to the country and fewer successfully thrive their, and the urban populations “make do” with fewer (surviving) children, successively.
Would people want to give up an urban life, even one that’s on the decline, and would they be able to pick up the skills and land for the country, and when it’s likely that the land beyond the cities was still more violent and dangerous? If people did move to the country, I might’ve expected them to transmit some culture, maintain some literacy, and I’m not sure that happened.
Imagine an economic crunch today really dropped urban wealth and employment, and a increasingly fairly lawless rural periphery of the country, plagued by marauding foreign bandits, without much “diversity” or appreciation for cultural outsiders joining local communities and a fair amount of suspicion of the urban core. Then that goes on for decades and centuries.
How many people would really move out of the cities, become migrants to the country and how many would just stay put and downsize and try and keep up some semblance of their urban predecessors lives? The culture of Rome, being a Roman, in particular, was I think even more focused on being a citizen, of the city, than our culture, and perhaps had less of a national culture bridging the city and country.
The only Roman DNA we have at the moment is Roman Britain – IRC out of five, four samples that look essentially native Celtic in profile, and one that looks Levantine or Egyptian, with some African component. How representative of turnover is this? More or less in the border marches than the core? Proper Roman DNA from the densely settled parts will matter.
“The only Roman DNA we have at the moment is Roman Britain.”
“four samples that look essentially native Celtic in profile”
If this is true then how is it reconcilable with what Mr. Cochran wrote above? I admit to being uninformed about this subject, but I’m fairly certain present-day Italians don’t resemble Celts.
I assume he is speaking only for Rome and Italy, and the cities there, and that Roman Britain would be a different case.
“If people did move to the country, I might’ve expected them to transmit some culture, maintain some literacy, and I’m not sure that happened.”
The collapse of literacy and of the classical Mediterranean trading economy have been blamed (controversially, but plausibly) on the Islamic conquests of the Southern and Eastern shores of that sea, with the loss of the papyrus trade from Egypt.
46.6K insulae would give well over 1 million inhabitants if we look at the density of modern slum tenements. I don’t have scholarly authority, but I have been in a number of them, and they can easily have 40+ occupants in a multi-story building….Pretty amazing that Rome could feed such a population.
It fits with what we know of grain imports. Probably.
There were two engines of diversity in Rome in the centuries after Christ. First there was the effect of legionary duty. If you were a free Roman peasant you were subject to serve in the legions for twenty years. For many that was a life’s sentence. After the great expansions of the Caesars these tended to be garrison troops who served on Hadrian’s Wall or the Limes.
If you were on Hadrian’s Wall you probably married a local girl and spread your Italian genes into Northern Europe.
If you were one of the one third of Rome who was brought to the city as a slave you might be manumitted. Wealthy Romans would buy talented Greeks or similar slaves who would be promoted into Roman businesses as artisans. You as their master and patron would set them up in business, providing the storefront where the former slave and his family would dwell. As their patron you shared in the proceeds. So it became a common business practice for wealthy Romans to proliferate small shop keepers and restaurateurs.
After a while the former slaves would be manumitted. This meant that their children were also free and were often given Roman or Roman sounding names. So after a while a large proportion of nominal Romans were no longer Italians but were Syrians, Gauls, Greeks, and others. The newly manumitted slaves all spoke Latin and had Roman sounding names. It became hard to tell.
So Roman citizens gradually became non-Italians while on the frontiers the real Italians were also diluting their genes. Without meaning too the Romans had set up two ‘melting pots’.
You’d think so, but looking at the genetics, there’s no sign of either trend.
This suggests the populations in those melting pots resisted being molten down, or the molten ones had low fertility and vanished.
Of course, we could be looking at two sides of the same coin. If mixing reduces fertility, resistance to mixing can evolve. There’s positive feedback too: Once you have reproductive xenophobia for any reason, mixed offspring is affected by it.
Going back to the previous post’s topic, if outbreeding can reduce fertility (I know you don’t believe it), the Ashkenazi population bottleneck may have been just that: Reduced fertility until they condensed down into into a new homogenous population.
It seems like the least objection with that is that there are a fair few groups with a more admixed history (at least, more diverged ancestors) that don’t have the same collapse after the initial period of admixture.
The Ashkenazi Jews started up in the Rhineland. They moved there. Not too many at first: that’s the bottleneck. The timing fits the genetic facts
Rome had high mortality: crowd diseases plus malaria by the time of the Empire. Any population in Rome was going to dwindle away without a constant reinforcing flow. Having the City sacked multiple times in the late Classical period probably didn’t help.
If low fertility was a natural consequence of hybridization between recently diverged lineages, it would be observed in many situations where it has not been. Doesn’t happen.
It appears that male slaves rarely reproduced and female slaves had such a low fertility that after three children she and the children were manumitted (Columella). By contrast, in the Soviet Union you needed nine children to be awarded the Maternal Glory medal (“Материнская слава”).
Taking a rural resident from northern Europe and placing him into classical Rome where there were al kinds of diseases he had absolutely no genetic resistance to explains why they died out rather than pass on their genes to modern day Italians. All cities before 1900 were population sinks but the survival rate of northern European slaves in Rome during this period must have been especially low. It doesn’t surprise me that there is little signature in modern populations of the multitudes of slaves imported to classical Rome. The incredible population density of classical Rome, the lack of clean water for the poor, the nasty diseases they had no resistance to, plus their was always an influx of new slaves captured in war to replace the ones that died. Julius Caesar sold at one time 53,000 slaves to a slave wholesaler, basically the entire population of a conquered region of Gaul. Talk about flooding the market, the price of a slave must have been cheap that year.
if cities were a population sink and prosperous farmers in the surrounding countryside were the main source of surplus offspring then even if Rome was 1/2 non-Italian in 400 AD they might have all been gradually replaced in the following centuries.
also soldier’s by blows generally have a particular easy time of it so i don’t think soldiers have much effect expect maybe the 20 year colonia men with a farm and a wife who had kids after retiring.
meant to be “generally don’t have an easy time”
So, where did the Natufian/Bronze Age Levantine/Iranian neolithic like admixture in southern/central Italians come from(because of course we now know south/central Italians cannot be a simple mix of Balkan/Anatolian farmers/Indo-Europeans/WHGs, both autosomally and based on y-dna) if not Roman immigration? Some bronze age migration?
Couldn’t find an email for you Greg, but is it even worth it to pursue an education/career in population genetics with the way the university system is now?
Do you have a reading list of textbooks that would be beneficial to someone unfamiliar with population genetics but has an undergraduate understanding of biology / genetics? A list of history books would be appreciated too.
Thanks.
The big expansion of slaves into Rome happened soon after the defeats of Carthage and Corinth. Presumably the slaves would have come from Greece and North Africa as a result. It was the flood of cheap slave labor that upset the balance between large and small farmers that many argue was what led to the end of the Republic. That suggests the slave population was dispersed throughout the peninsula.
As far as after Rome expanded into the rest of the Europe, I’m not sure I’d assume the Romans brought high numbers of slaves onto the peninsula. They tended to resettle barbarians in the provinces, not the peninsula. It’s a good question though.
I suspect the genetic destiny of slaves was very uneven. Male slaves, who were probably the bulk of them, I’m sure were reproductively unsuccessful overall, though some high-skilled slaves may have done well. I expect female slaves to have been universally used as concubines, and probably contributed to the ongoing gene pool.
The concubines contributed in a way that has left little to no trace.
“used as concubines, and probably contributed to the ongoing gene pool.”
long term reproductive success proportional to class imo
ex-legionary -> farmer (middle class)
ex-slave -> free washer woman with one of master’s kids (underclass)
The cities were population sinks,
This is the real reason. You even do not need empire collapse to have foreign population sink away. Example: Amsterdam, Xian.
Middle eastern traders (Arabs) had been living in Amsterdam since medieval times. Central Asia traders even had their districts or neighbor in Changan (Xian) in Tang dynasty. But at end, these foreign settlers were always replaced by the native people at end since urban people were always below their own replacement level.
It is rural people who are always produce surplus offsprings. Even today, traditional large families in USA are always found in countryside.
“It is rural people who are always produce surplus offsprings. Even today, traditional large families in USA are always found in countryside.”
Generally, this is true. In the countryside, children are cheap, helpful, and the older ones can mostly take care of younger ones. In a city, children are expensive, distracting, and require adult supervision at all times.
I would live to have 10 children, but that would require me to drive a bus, rent four apartments, and have a wife who doesn’t work.
Living in the city with only two kids, I have no ability to grow any of my own food, nor do I have room or time to repair anything myself.
Meanwhile, my country cousins own huge plots of land, and have at least four children each.
If anyone wanted intelligent people have more children, all they would have to do would be to build all new scientific institutes very far out in the countryside, and pay post-docs and professors there enough to support a large family on one salary.
Might be hard to tell with the Greeks. Their genetic similarity to Italians is uncanny according to some https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2730349/
To get good numbers on this, wouldn’t you need samples of Italians from now and from back then? Ancient Romans, pre-conquest would serve as a control group, and if modern Italians were no more similar to Spaniards or Germans or CEU than ancient Romans were then you’d be right. But you’d want to take care in selecting a population! Northern Italy has a higher frequency of Germanic y chromosomes such as I1 and tends to cluster a bit closer with other Europeans on a PCA graph. Lombards and Ostrogoths could’ve done there what you believe the slaves failed to do in Rome: contributed genes.
Pingback: Demography: How to Make America White Again | whyteablog