Back to Post-publication discussions

Intelligence Trends in Ancient Rome: The Rise and Fall of Roman Polygenic Scores

Submission status
Reviewing

Submission Editor
Noah Carl

Authors
Davide Piffer
Edward Dutton
Emil O. W. Kirkegaard

Title
Intelligence Trends in Ancient Rome: The Rise and Fall of Roman Polygenic Scores

Abstract

We analysed 127 Ancient Roman genomes with a view to understanding the possible reasons for the fall of the Roman Empire. Taking the polygenic score for educational attainment (EA4) as a proxy for intelligence, we find that intelligence increased from the Neolithic Era (Z= -0.77) to the Iron Age (Z= 0.86), declines after the Republic Period and during the Imperial Period (Z= -0.27) and increases in Late Antiquity (Z= 0.25) and is approximately at the same level today (Z= 0.08). We show that this is congruent with a cyclical model of civilization based around intelligence, with the documented history of Rome, and also with patterns of immigration into Rome.

 


 

Keywords
gwas, polygenic scores, Roman empire, Civilization Cycles, Rome, ancient Rome, Cold Winters Theory

Pdf

Paper

Reviewers ( 1 / 0 / 3 )
Reviewer 1: Accept
Reviewer 2: Reject
Reviewer 3: Accept
Reviewer 4: Accept

Tue 25 Apr 2023 11:14

Author

I have responded to the comments by reviewer 4 (submitted by e-mail) and uploaded the edited version.

  1. Changed the wording as suggested.
  2. Changed the sentence as suggested.
  3. Done.
  4. Moved the cyclical theory of civilization a few paragraphs below.
  5. Added that it has been observed in developing countries too.
  6. Added “in the Mediterranean region”.
  7. “we cannot simply assume that these migrants had lower polygenic scores at the time of the Roman Empire”. No, we cannot assume this, but it is a possibility worth taking into consideration. That is why we wrote “the average IQ of such people may be lower”.
  8. Changed to “During the Imperial period, ancestry in the city of Rome shifted toward the eastern Mediterranean”.
  9. Replaced “empires” with “civilizations”
  10. Corrected the spelling.
  11. I removed the “double-relaxation” and just focused on the fertility dynamics.
  12. Replaced “empire” with “Roman civilization”.
  13. Changed to “Cox et al. (2022) used a polygenic score derived from modern samples to predict skeletal stature under ancient DNA, predicting 6.3% of femur length of ancient skeletons.”.
  14. Changed “European ancestry” to “Western European” ancestry.
  15. Added title and legend to table.
  16. Remove the sentence referring to the rate of innovation declining after 200 BC.

Replying to Forum Bot

Authors have updated the submission to version #5

 

Bot

Authors have updated the submission to version #6

Reviewer

I found only one thing that needs to be corrected. Look at the top of page 6. The periods are defined in a very strange way. This needs to be clarified.

Author

These are the periods are defined by the original publication in Antonio et al..when they do not include the entire historical period it is because there were no ancient genomes recovered from that time, hence the gaps (i.e. from 200 BC to 27 BC).

Replying to Reviewer 4

I found only one thing that needs to be corrected. Look at the top of page 6. The periods are defined in a very strange way. This needs to be clarified.

 

Reviewer

Thank you for the revision, now accept.

 

Replying to Davide Piffer

Replying to Reviewer 3

I have read the paper, thanks.

Brief suggestions:

Add ds (or zs) and IQs in abstract and Figure 1 (or make new Figure 2).

E.g.,

old: "we find that intelligence increased from the Neolithic Era to the Iron Age, declines after the Republic Period and during the Imperial Period and increases in Late Antiquity and is approximately at the same level today"

new (numbers invented by me): "we find that intelligence increased from the Neolithic Era (representing z=0 or IQ 85) to the Iron Age (z=1 or IQ 100), declines after the Republic Period (z=0.50 or IQ 92.5) and during the Imperial Period (...) and increases in Late Antiquity (...) and is approximately at the same level today"

Explain what is the standard for IQ 100 (natives in Britain today? Italy today?).

Best and thanks

Thanks for the comments. I have replaced figure 1 with a figure showing individual Z values. I have also added a table (table 1) showing the average Z score by period. I refrain from performing a simple transformation to IQ because the conversion is not straightforward (PGS only includes additive genetic variance and represents only a fraction of it). However, I believe the corresponding change in IQ SDs is pretty similar to the change in PGS SDs but it would be beyond the scope of this paper to calculate IQs.

I have edited the abstract by adding the effect size (Cohen´s d) of each period (with pre-Iron Age as the reference).

 

 

Author
Replying to Reviewer 1

The Introduction is still too long. In fact, it's just as long as it was in the first version. Maybe the authors think I will forget this point if they just ignore it.

The Introduction is still an exercise in "putting the cart before the horse." The authors try to explain their findings before they actually present them. Such an approach would be defendable if the same findings had appeared in a previous study. But there is no such previous study, certainly not the massive work by Edward Gibbon. He never claimed that mean cognitive ability had declined during Rome's Imperial period.

I would be more forgiving if the authors could write clearly and succinctly. Unfortunately, that doesn't seem to be the case. I'm still having trouble understanding their argument, since it branches off in several directions. Here is my understanding of what they mean to say:

1. During pre-Imperial times, the Romans had "high intelligence and other pro-social psychological traits." The authors seem to be arguing that pro-sociality creates an orderly environment that is conducive to pursuits with high cognitive demands (cf. Gregory Clark).

2. The establishment of the Empire led to a decline in pro-sociality. Life became too good. There was "clean water, basic medical care and an abundance of food" (Oh really?). There was also immigration from the Greek-speaking eastern provinces. All of these trends presumably caused a decline in mean cognitive ability. I say "presumably" because the authors provide no evidence that the Greek-speaking East was less intelligent on average than the Latin-speaking West. The stereotype of the time was the reverse: the East was the home of thinkers and philosophers. In any case, the authors should not assume what they wish to prove.

3. Meanwhile, there was climate warming. Perhaps this was a factor in Rome's decline, or perhaps not. The authors are like the hunter who tries to catch two rabbits at the same time.

4. With the decline of the Empire, the Roman population may have entered "another eugenic phase where intelligence is rising." I initially saw this as a reference to Christianity and selection for pro-sociality (cf. Schulz et al.), but apparently I was mistaken.  

I'm coming around to the other reviewer's (harsh) judgment. The results are interesting. Unfortunately, they are sandwiched between a lot of intellectual musing and meandering.

Thank you for your feedback and suggestions. I have taken your comments into consideration and made the necessary revisions to address the issues raised. I have extensively re-written the introduction to make it clearer, and I have removed the speculative parts of the introduction, including the reference to the Cold Winters theory and the assumption about immigration from the Eastern Mediterranean causing a decrease in polygenic scores. I agree with your point that there is no evidence to suggest that the Greek-speaking East was less intelligent on average than the Latin-speaking West.

Furthermore, I have removed the paragraphs describing the ancestry shifts from the introduction, as you recommended. However, I respectfully disagree with your statement that "The Introduction is still an exercise in 'putting the cart before the horse.'" In fact, the hypothesis that there was a cognitive decline in classical antiquity has been proposed before and is not merely a post-hoc interpretation, as you suggested.

Additionally, you mentioned that there is no previous study with similar findings to support our approach. While it is true that the findings presented in this paper are novel, the hypothesis itself is not. Peter Frost, in a blog post published around the same time our paper was submitted, discussed the decline in mean cognitive ability during Classical Antiquity and proposed potential causes based on historical phenomena discussed by other authors.

To support this, I have included a reference to Peter Frost's blog post in the introduction, where he discusses the decline in cognitive ability and suggests three main causes. This demonstrates that the hypothesis of a cognitive decline in the Mediterranean has been previously advanced by Woodley of Menie et al. (2019) and Peter Frost in 2022.

Thank you once again for your valuable feedback, which has helped improve the clarity and accuracy of the introduction.

Bot

Authors have updated the submission to version #7

Reviewer
Replying to Forum Bot

Authors have updated the submission to version #7

Peter Frost is a marginal scholar of little importance. More to the point, a blog post is not peer-reviewed. It belongs to the "grey literature" that some people consider to be acceptable and others do not.

Nothing is forbidden in a scientific paper. It's not forbidden to cite blog posts or use words like "dysgenic," but these are things that will cause some people to reject your findings without giving them a fair hearing.

Other than that, I have only one remaining criticism.  On page 2, you write "intelligence ... is highly selected for under harsh conditions, where complex problem-solving skills are crucial (Lynn, 1987)." As I understand Lynn's argument, it's not harsh conditions per se that select for intelligence but rather harsh conditions that occur in a predictable manner (and which can thus be alleviated through the use of problem-solving skills). 

In any case, I approve this manuscript for publication.

Bot

The submission was accepted for publication.

Bot

Authors have updated the submission to version #9

Bot

Authors have updated the submission to version #10