Few Foreigners in Ancient Rome

September 17, 2010

A common narrative is that the Roman Empire fell because Italy was overrun by (mostly non-European) foreigners who replaced the native population. But according to historian David Noy, the number of foreigners in the city of Rome — which would have been much higher than anywhere else in Italy — was only about 5% at its peak. And these foreigners (especially the slaves) had higher mortality rates and lower birth rates than natives, and were sometimes even subjected to mass expulsions.

In addition, half of the main areas from which they came were other parts of Europe adjacent to Italy: Gaul (France), Hispania (Spain and Portugal), Central Europe (Southern Germany), Eastern Europe (the Balkans), and Greece. And even those from outside of Europe were predominantly of Greek descent, while the Jewish ones never fully assimilated into Roman society, remaining an identifiable minority.

Population Size:

Immigration was essential to Rome both demographically, to increase or at least maintain the size of the city's population, and socially, to provide skilled workers and soldiers. The slave trade met some of the requirements, but free immigrants were always needed. Provincials probably began to outnumber Italians among newcomers to Rome in the first centuries BC and AD. The third century AD, when all recruitment for the Praetorian Guard was done in the provinces, may have seen the numerical peak of Rome's foreign population. It is plausible to suppose that at least 5% of the city's inhabitants were born outside Italy in that period; the reality could be much greater.


Foreigners who did not have Roman citizenship were always liable to summary expulsion from the city, and by the fourth century the possession of citizenship was no longer protection against such treatment. Although there was a certain amount of xenophobia within the Roman literary class, expulsion was only used in certain circumstances: to deal with the actual or potential misdeeds or alleged bad influence of specific groups (which could be defined by nationality, religion or occupation), or to counteract the effect of food shortages by reducing the number of mouths to feed. Expulsions were probably not carried out very efficiently, and were always short-lived.

Mortality Rate:

It is generally agreed that mortality was probably higher in Rome than elsewhere in the Roman world, because of insanitary living conditions and the risk of contagious diseases; diseases such as tuberculosis may have been endemic. Newcomers to [17th-18th c.] London were more susceptible to plague than natives were, and the same point has been made about the greater susceptibility of Rome's immigrants to plasmodium falciparum malaria. Tuberculosis might be particularly dangerous to the young adults who probably formed most of the immigrant population. [...] Slaves are likely to have suffered from higher mortality than the free population, and immigrant slaves would have been particularly vulnerable to diseases which were not prevalent in their homelands.

Birth Rate:

It is also likely that the birth rate would have been lower at Rome than elsewhere. Many migrants coming to the city would already have spent some of their fertile years elsewhere, and the slave part of the population would have been less fertile than the rest. Free male citizen immigrants may have postponed marriage until they had access to the corn dole, which from the time of Augustus was only available to a restricted number of recipients. In London, for similar reasons, the natives were closer to reproducing themselves than migrants were, and the same would almost certainly have been true for Rome.

Asia Minor:

Although literature emphasizes the significance of Asian slaves at Rome, inscriptions present a rather different picture. The large number of epitaphs in Greek, especially for people from the province of Asia, is consistent with the large number of recorded peregrini [foreigners] in suggesting that the migration of people of free status was particularly significant for this area. The evidence is, however, almost exclusively concerned with the Greek population of Asia Minor, and there is very little sign of people of non-Greek background coming to Rome except as slaves. This is consistent with the general predominance of the most romanized/hellenized section of their home society among free migrants to Rome.


However, most Syrians arrived at Rome through the workings of the slave trade. Syrus was a common slave name, although not necessarily given only to Syrians, since the association Syrian = slave seems to have been very widespread....

Voluntary migration from Syria to Rome would probably have begun in the late Republic. Most of the evidence, however, is from the second century AD or later. There is a clear implication that some of the slaves and ex-slaves labelled Syrians in the literary sources were thoroughly imbued with Greek culture, whether their ancestry was Syrian, Greek or mixed. Solin (1983, 722) notes that Syrian immigrants in general tended to be of Greek descent or at least to be from the most hellenized part of Syrian society.


Most references to Egyptians at Rome concern Alexandrians, apparently of Greek extraction, rather than 'indigenous' Egyptians. On the other hand, the stereotyped Roman image of Egyptians concentrated on the aspects of their behaviour perceived as most outlandish, particularly the worship of animal-gods, and largely ignored the Greek component of their culture. There seems to be something of a contradiction between image and reality which may be due at least in part to anti-Cleopatra propaganda and its legacy.

North Africa:

North Africa contained some cities which were Greek, Libyan or Phoenician foundations, but many of the main population centres began as Roman colonies (notably the re-established Carthage) or military settlements. Ricci (1994b, 198) believes that the colonization programme of Julius Caesar and Augustus in North Africa also stimulated a population flow from there to Rome. The inhabitants of the area came from a wide variety of ethnic backgrounds (Italian, Greek, Punic, Libyan, Berber, Jewish), but, as with other areas, it is likely to have been the most romanized/hellenized section of the population which provided most of the free migrants to Rome.


The group which made the greatest effort to retain a separate identity was the Jews. In their religious and communal institutions, their use of separate catacombs, their epigraphic and liturgical use of Greek, and even their naming practices, they behaved differently from others and were able to pass on a Jewish identity, so that people whose ancestors had lived at Rome for generations and who were otherwise well integrated into Roman society were still identifiably Jewish.

David Noy. Foreigners at Rome: Citizens and Strangers. London: Duckworth, 2000.

Related: Biochemistry of Skeletons from Ancient Rome


Anonymous said...

This blog isn't much more than an ethnic-Mezzogiorno public relations effort. However, this post seems more plausible. The north Africa of Roman times, with no Arabs and probably no blacks outside of Egyptian Nubians, was very different than the north Africa of today (with much Arab and sub-Saharan DNA). The numbers of foreigners in ancient Rome has always probably been exaggerated, and probably left only a small genetic imprint.

Anonymous said...

I do get the feeling, as the world goes more left wing it's becoming more anti-Italian. If you read Berkhardt, North Italy always related to south Italy in the old days, and considered people North of the alps as barbarians. Even Machiavelli dreamed of a United Italy, so I don't understand this North South divide, & the North pretending their part Barbarian & proud of it(!?) But I do hear that even in sub-saharan Africa they relate to Germans & put down Italians. Just read car reviews from Africa, Alfa Romeo are built by Godfathers & Pizza makers. Something strange is going on.

Anonymous said...

Stop all this special pleading.

Tenney Frank was right and modern genetic studies show how Levantine Italians are,especially South Italians and Sicilians.

Italianthro said...

Tenney Frank's "Orientalization" hypothesis was refuted soon after it was published, and modern genetic research confirms that refutation, as well as more recent historical evidence, by showing that Levantine slaves/foreigners left no significant mark on Italy's gene pool.

Anonymous said...

What refutation.
Eurogenes and Dodecad show plenty of West Asian and SW Asian in Italy, especially the mainland South and Sicily.

Italianthro said...

Those genetic components are present throughout Europe in a clinal pattern that reflects prehistoric demographic events and geographical distance. They have nothing to do with the subject of this blog post.

Anonymous said...

So we are expected to believe that Roman Italy was not affected by a total of millions of slaves over the centuries, many of whom were freed.
Their children became Roman citizens.

Italianthro said...

As Noy explains, slaves (esp. foreign slaves) had much lower fertility and higher mortality rates than natives. Also, according to Nathan Rosenstein, the "millions of slaves" estimate is incorrect, and urban population increases were caused by migrants from rural areas of Italy, not slaves or foreigners. Indeed, there's no evidence for a large-scale infusion of new genetic material into Italy at that time.

All of this has already been covered. Stop spamming my blog with your uninformed opinions.

Anonymous said...

That's only Rosenstein's opinion.

He has no scientific proof of a very unlikely proposition.

Italianthro said...

I suggest you learn the difference between "opinion" and "research". Professor Nathan Rosenstein, Ph.D is "a specialist in the history of the Roman Republic and early Empire", and his book is a work of historical research that cites multiple sources as proof. Although even his casual opinion would be infinitely more valuable than yours.

Anonymous said...

Tenney Frank's studies of ancient Roman inscriptions are old but exhaustive and any casual observer in Italy,and Sicily and Southern Italy above all, would see an Italian-Syrian/Lebanese phenotypical link.

Italianthro said...

>>> "Tenney Frank's studies"

...have been refuted.

>>> "an Italian-Syrian/Lebanese phenotypical link"

That "link" (actually overlap between different Alpine, Dinaric and Mediterranean variants plus similar climatic adaptations) dates back to at least the Neolithic and can be seen in more places in Europe than just Italy. It has nothing to do with the subject of this blog post.

Unknown said...

What do you think of the views of Theodor Mommsen?