Talk:Demography of the Roman Empire

Move discussion in progress
There is a move discussion in progress on Talk:Demography of Birmingham which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 13:29, 3 April 2015 (UTC)

Table from the wrong era
The table at the end of this article showing the populations of various cities appears to be from the wrong era. According to the citation, the population data is for the 11th century, long after the imperial period discussed in the rest of the article. I think it should be removed. Ideally, it would be replaced with a table for the appropriate (1st to 2nd century) era. — Preceding unsigned comment added by Aiwendil42 (talk • contribs) 19:30, 30 January 2016 (UTC)
 * To judge by the note, it's for the right era but is a re-used estimate originally from J. C. Russell, Late Ancient and Medieval Population (Philadelphia: American Philosophical Society, 1958), 65–83. I would leave it for the moment, but if anyone could check on the original reference that would be great. A more up to date one would be even better, and a statement of whatever academic consensus may currently exist would be perfect. Richard Keatinge (talk) 11:59, 1 February 2016 (UTC)

"I would leave it for the moment" because clearly, completely garbled information is better than no information. Rome had 350,000 inhabitants neither in the 1st century nor in the 11th century. The population of Rome would have been near 350,000 in the late 5th century, on the eve of the collapse of the Western Roman Empire (History_of_Rome). If you think about it, this makes sense, it is a hint that the reference (Russell 1958) has the title "late ancient and medieval population". --dab (𒁳) 08:13, 24 July 2016 (UTC)

Some problems with the population table
The main table of population data in the article has a few issues. My main gripe is with the areas given. I've looked at the source (Frier's "Demography" section in The Cambridge Ancient History XI: The High Empire) and he has taken the data from a 1973 study, which is itself based on Beloch's 1886 book (Die Bevölkerung der griechisch-römischen Welt). Unfortunately I haven't been able to find an English translation of this book, so I can't know for certain exactly which areas Beloch was referring to. However if you add up the areas of the modern countries listed, they do not match up with the area figures given in the table. For instance in Note 4 it says that the region termed the "Greek peninsula" is: "Defined to include the modern territories of Greece, Albania, and European Turkey", and says this area is 267,000 square kilometers. This cannot be correct. If you add up the modern areas of these places; Greece (131,957 km2), Albania (28,748 km2), and Eastern Thrace (23,764 km2), it is only 184,469 km2. So where's the rest? I think the answer is that European Turkey was a lot bigger in Beloch's time. It included Eastern Rumelia for one (about 50,000 km2) and you also have to include the Republic of Macedonia (at 25,713 km2). You might also need to include all of Kosovo, a part of Serbia, or more of Bulgaria to get the area up to the cited figure of 267,000 km2. So, what I'm saying is that the quoted area in the note, as being only modern "Greece, Albania, and European Turkey", is clearly wrong. Beloch was either using the Administrative divisions of the Ottoman Empire in his time (which only partially link up with modern borders), or (perhaps more likely) he was using the borders of the ancient Roman provinces, which also do not match up neatly with modern borders. The cited figure seems equivalent to the Roman provinces of Achaea, Macedonia, Epirus and Thracia (like in ). Likewise the note about the region called "Greater Syria" being "Defined to include the modern territories of Syria, Lebanon, Israel and Palestinian territories" is also clearly wrong, because modern Syria on its own is 185,180 km2 (it is clearly only talking about the costal Levant, which includes only the western half of modern Syria). So I think the notes are going to have to be rewritten. There are other problems, like the fact that it doesn't give detailed figures for the "Annexations". For instance, the quoted area of the whole Empire as 3,339,500 km2, is only valid for the first set of numbers from 14 AD and doesn't include the annexations by 167 AD. We would need a separate area figure for that, given that we are given a population density. In fact you can derive this number easily by dividing the 167 AD population figure of 61.1 million by the cited density of 15.9 p/km2, and it is 3,861,600 km2 (and increase of 522,100 km2). I think I will try to improve the table somewhat, although it may not be easy. --Hibernian (talk) 09:17, 1 January 2017 (UTC)

Removed section that misrepresents sources
I removed the section below due to issues with misrepresentation of sources.
 * Conversely, the foreign population of Rome was very small. Scientific evidence has shown that non-locals were in the minority, and most came from other parts of Italy or nearby provinces in Southern/Central Europe. Biochemical analysis of 166 skeletons from three non-elite imperial-era cemeteries in the vicinity of Rome revealed that only 1 individual definitely came from outside of Europe (North Africa), and another 2 possibly did, but results are inconclusive. Other than that, there was no apparent net inter-regional migration in the imperial period, except perhaps a small continuous resettlement of Easterners in the West.

Quoting from elsewhere:

I think these sources would still be useful contributions to the article, but we should rewrite to make sure we reflect their findings accurately. Ganesha811 (talk) 19:36, 18 December 2019 (UTC)

Just read the articles, instead of listening to some random dude on reddit. I've read again the articles and the numbers are correct. I've modified the page because the isotope analysis can only inform us about the birthplaces, and not the ancestry, unlike genome wide analysis. If you want to add another section about genetics in this page quoting Antonio et al, do it, but the actual paper also tells us that the population of Rome was wiped out by multiple plagues and barbarian invasions and the area was resettled by people from the North. LambdofGod (talk) 15:23, 23 December 2019 (UTC)
 * Hi! I've just finished re-reading the papers (here and here) and I'm not sure why you think the article quotes them accurately. The article currently says:


 * But the actual journal pieces say different. "All Roads Lead to Rome" says that there are four individuals with "clearly anomalous isotope ratios that suggest an Alpine or Tyrrhenian origin, and four more "whose isotopes may indicate they were immigrants." That's 8, not 3 as the article currently suggests.


 * This gets to a larger point, which is that from these sources, we cannot say that "Conversely, the foreign born population of Rome was very small." This is an extremely broad and strong statement which apparently covers all time periods of Ancient Rome based on two studies which themselves warn not to draw broad conclusions from them. The scientists, sensibly, state that these results are only based on particular cemeteries at particular times and do not cover all possible immigrants or types of immigration.


 * The other article is also not accurately represented. It says that:


 * I think it is clear that the article as written above does not accurately reflect, in number or in character, what the studies say about foreign-born individuals in Rome. It draws broad conclusions from small samples of specific data and reduces ⅓ of one population studied and 8 individuals in another population studied to "only 1 individual definitely came from outside of Europe (North Africa), and another 2 possibly did, but results are inconclusive." This is not correct. I will remove the material above from the article. If you would like to seek dispute resolution to resolve this or an RfC, go for it. Happy to keep discussing this with you. Ganesha811 (talk) 16:46, 26 December 2019 (UTC)

There are 166 samples from 3 locations in those 2 studies and that's not few. Stop with this nonsense. The data are correct. If you think that those samples are not "representative" than please show us other studies with a much larger number of samples with different results. LambdofGod (talk) 23:29, 26 December 2019 (UTC)

I've edited the article to make it cloder to the article.

About the first study:

Alpine or Tyrrhenian origin are still in Southern Europe.

About the second study:

"We hypothesize that migrants most likely came from higher elevations to the East and North of Rome. One individual with a higher δ18O value may have come (as a child) from an area isotopically similar to North Africa"

"There are several possible origins for the outsiders (at birth) buried in the cemetery of Isola Sacra who must have come from regions where δ18O of local precipitation is lower than in Rome by up to 2.6%. One possible region of origin is the Roman Imperial provinces lying to the North of the Italian landmass."

"As another possible locus for the outsiders, we note that δ18O values of modern meteoric water vary continuously to values up to 4% lower than those encountered in Rome at distances as close as 100 km in the foothills and heights of the Apennine Mountains (Longinelli and Selmo, 2003). Derivation of the outsiders principally from this region seems to be the most likely scenario. Other possible regions of origin of the outsiders might be the Iberian Peninsula or Greece, both of which were under Roman control at this time."

" The continuous gradation of δ18Oap between local and outsider δ18Oap values suggests that these individuals came from locations at gradually farther distances and gradually higher elevations than Rome. If the outsiders were from as far away as southern Gaul (where δ18O of rain is about 2% lighter than in Rome),"

" The individuals with low δ18Oap values could have been from as close as 100 km to Rome, in the hills surrounding the Apennine Mountains. It is also possible that they came from the transalpine provinces of the Roman Empire much further to the north, where low δ18Oap precipitation falls even at low elevations. However, due to the observed scatter of the δ18Oap data around the "local" Roman range, we conclude that this latter explanation seems less likely. Only one individual was found to have a conspicuously high δ18Oap value; the observed value is consistent with an origin in a region with higher δ18Oap in drinking water, like the Nile Delta, although it is impossible to exclude possible origins in southern Italy. "

LambdofGod (talk) 23:49, 26 December 2019 (UTC)
 * The two references are primary sources which makes them WP:UNDUE and unsuitable for use in Wikipedia's voice. The conclusion is undue unless a secondary source has verified it. Johnuniq (talk) 06:20, 27 December 2019 (UTC)
 * Those quotes don't change the basic issues here. Not only is this WP:UNDUE as pointed out by the user above, but the whole section is framed in such a way as to mislead the reader. We cannot say the "foreign-born population of Rome" was very small (which implies this is the case for the entire history of Rome) on the basis of two studies with a couple of hundred graves each, especially when the authors of the studies themselves warn us against this! The phrasing "foreign-born locals" is also confusing.


 * Using the modern definition of Europe as a benchmark for foreign-ness in Ancient times is also an issue - the Romans would have considered a Greek or a Gaul just as foreign as a Numidian or a Jew. At many times in earlier Roman history (which the article should cover too, after all) they would have considered anyone born outside the city of Rome itself a non-citizen and therefore a foreigner. The key issue was Roman citizenship, which was not extended to all men in the Empire until 212 AD, well into the Imperial period. And even then the concept of "foreigner" is culturally defined. The whole section needs a much more comprehensive and subtle treatment than it has now. Particularly strange is the phrasing saying that "other than that," there was no apparent inter-regional migration, which implies that we should view the individuals in these studies as the only evidence of migration, leaving aside the extensive written and other evidence of migration!


 * I'm removing the section again. We cannot draw the conclusions the section does on the basis of these studies and to include them would only serve to mislead the reader. Ganesha811 (talk) 15:28, 27 December 2019 (UTC)
 * To add context, Elio lo Cascio in "The Impact of Migration on the Demographic Profile of the City of Rome: A Reassessment" (2016) writes:
 * "It is worth noticing that the corpus of inscriptions referring to the foreigners at Rome led David Noy to estimate the number of immigrants coming to Rome from outside Italy conservatively at only 5% of the urban population (although this was conceived of as an absolute minimum figure)... one cannot refrain from underlining the importance of the finding that migrants still constituted a very high proportion, perhaps even one third, of Rome's population during the principate."
 * So this review of contemporary research suggests that between 5-33% may be reasonable estimates for the number of foreigners in Rome during the early Empire. This sort of secondary source info is the sort of that that should be included, not broad conclusions of original research drawn from a two very specific studies. Ganesha811 (talk) 17:51, 27 December 2019 (UTC)
 * So this review of contemporary research suggests that between 5-33% may be reasonable estimates for the number of foreigners in Rome during the early Empire. This sort of secondary source info is the sort of that that should be included, not broad conclusions of original research drawn from a two very specific studies. Ganesha811 (talk) 17:51, 27 December 2019 (UTC)
 * So this review of contemporary research suggests that between 5-33% may be reasonable estimates for the number of foreigners in Rome during the early Empire. This sort of secondary source info is the sort of that that should be included, not broad conclusions of original research drawn from a two very specific studies. Ganesha811 (talk) 17:51, 27 December 2019 (UTC)

Outdated data
I see huge space devoted to data from sources over 60 years old, while we have brand new population estimates for regions and cities. For example, I was just reading a 2017 book with a population total of 75 million based on the newest estimates, with a detailed breakdown by region and with population density as well. While we have Hanson's database of city sizes (in area) and his population estimates. RafaelG (talk) 03:11, 11 November 2021 (UTC)
 * go for it! Ganesha811 (talk) 05:04, 11 November 2021 (UTC)

Genetic studies indicate substantial immigration
I've not fully read the isotope studies, but a recent genetic study indicates that the population in and around the city of Rome during the republic was 10% Eastern Mediterranean or Near Eastern, and wholly 2/3 during the imperial era, indicating large immigration. I recommend those interested read the article. I will also read the isotope studies and may update the article at some later point. MrThe1And0nly (talk) 18:27, 1 November 2023 (UTC)


 * Go for it! —Ganesha811 (talk) 13:30, 7 November 2023 (UTC)
 * I did :) MrThe1And0nly (talk) 21:32, 9 November 2023 (UTC)

Move discussion in progress
There is a move discussion in progress on Talk:Demography of Belfast which affects this page. Please participate on that page and not in this talk page section. Thank you. —RMCD bot 02:37, 1 February 2024 (UTC)

Change name to "Demography of Ancient Rome"?
This would allow to treat the demographics of the entire history of Ancient Rome here, including the Republic. Considering that there's not a lot of info on the demographics of the Republic anyway, it's not going to have its own article anytime soon, and it won't be a lot of work to add it here. MrThe1And0nly (talk) 21:21, 11 April 2024 (UTC)

Calling Iron Age Romans "European"
Deusestlux The authors (Antonio et al.) who studied Iron Age Romans called them "Europeans." You want to call them "Spanish-like" instead, going by a study of Etruscans (Posth et al.). It's obvious that despite the Etruscan genetic affinity to Latins, Aiken's study takes precedence, for it pertains exactly to the given subject - it being Iron Age Romans, and not Etruscans. Aiken was well aware of where Iron Age Romans position on the PCA, and also aware of their proximity to Etruscans, and yet still refrained from calling them "Spanish-like." It may perhaps have to do with the fact that they position almost smack in the middle between modern Spain, France and Northern Italy, requiring a more general term than simply "Spanish-like." Or it may instead be due to their haplotype heatmap, which suggests Iron Age Romans were "Central & Northern European" instead. Whatever it is, they chose "European" for a reason.

Therefore, unless you find a more recent study of exactly Iron Age Romans, refrain from changing statements which quote studies on Iron Age Romans verbatim. MrThe1And0nly (talk) 17:51, 7 May 2024 (UTC)

Genetic studies
@Deusestlux and @MrThe1And0nly, I've looked quickly at the citations. Various questions arise, especially when trying to consider what Wikipedia can say about how Roman (or Italian, or empire-wide) demographics changed over time. NebY (talk) 19:18, 9 May 2024 (UTC)
 * Some seem to have sampled modern populations. Which ones examine ancient samples?
 * How closely do those date and locate their samples - how closely are death / burial dates known, and how explicit are the researchers about find sites?
 * Is there any discussion of how representative those ancient samples are – do they come from periods when cremation was rare; if taken from graves, are the burials typical or for example, atypically prestigious; if from cemeteries, are those cemeteries used by freeborn citizens, and freedmen and freedwomen, and slaves, or more selectively?
 * Which ones refer to other ones we're citing?
 * Are they all publishing new research results, or do some provide an overview or synthesis of previous research instead?
 * Have any been challenged?
 * Is our current phrasing, such as saying some general ancestry is "revealed" appropriate for small sample sizes with clear exceptions (such as in the article as it stands, "two out of six individuals" being exceptions)?
 * We seem to be treating some research as definitive in an attempt to provide a single definitive account. On what basis? At this point, should we even try?