Let’s Talk About Ancient Empires And Their History

Bharti Prasad
8 min readApr 27, 2021

Analysis of some fascinating data

Photo by Mustafa ezz from Pexels

A full 90% of all the data in the world has been generated over the last two years. The internet companies are awash with data that can be grouped and utilised. — As stated by Petter Bae Brandtzæg in one of his article in 2013

From the dawn of civilization to the end of the Cold War, we have far less data than we do over the last 30 years. The amount of data we can examine until the advent of modern statistics in the mid-1600s is infinitesimal. It gets worse if we go back before the Domesday book in 1086.

Fortunately, nothing prevents modern historians and scholars from producing new historical data.

Axial Age Dataset

For the analysis, I have used a relatively very small dataset, Axial Age Dataset, by Seshat: Global History Databank, “a large, multidisciplinary team of evolutionary scientists, historians, anthropologists, archaeologists, economists, and other social scientists from around the world.”

This research employed data from the Seshat Databank (seshatdatabank.info) under Creative Commons Attribution Non-Commercial (CC By-NC SA) licensing.Turchin, P., R. Brennan, T. E. Currie, K. Feeney, P. François, […] H. Whitehouse. 2015. “Seshat: The Global History Databank.” Cliodynamics 6(1): 77–107. https://doi.org/10.21237/C7clio6127917.Mullins, D., D. Hoyer, […] P. Turchin. Preprint. “Mullins, D., D. Hoyer, […] P. Turchin. 2018. “A Systematic Assessment of ‘Axial Age’ Proposals Using Global Comparative Historical Evidence.” American Sociological Review 83(3): 596–626. https://doi.org/10.1177/0003122418772567.

The Axial Age dataset tracks a variety of sociopolitical norms and their development across key areas in Afro-Eurasia. The specific scores for each sociopolitical norm for each date (varying time spans between 5300 BCE and 1800 CE in 100-year increments) within 10 NGAs (natural geographic area) were agreed upon by a group of experts and compiled into the dataset.

What is Axial Age?

The Axial Age (also known as the Axis Age) is the era when the great intellectual, philosophical, and religious systems that came to form subsequent human civilization and culture appeared, much of the inhabited world at about the same time. The period during which most of the main religious and spiritual traditions emerged.

Exploratory Data Analysis

Link to my GitHub repository for this project and code.

Link to my Kaggle Notebook for the same project.

Matrix to visualize the pattern of missingness in our data
Matrix to visualize the pattern of missingness in our data
Bar chart to visualize the missingness in our data
The bar chart gives us an idea about how many missing values are there in each column
Heatmap shows the correlation of missingness
The dendrogram allows us to more fully correlate variable completion, revealing trends deeper

As we can see from the visualizations above, this dataset has a lot of missing values. Let’s jump right into data visualization since I’ve cleaned up the data.

Autocorrelation plot

The time periods covered by each NGA in the dataset are different, ranging from 5300 BCE to 200 CE. We can note that from 5300 BCE to 4000 BCE there were zero positive cultural feature observations. The general cumulative trend elsewhere was generally positive.

Most features continue to show a general upward trend over time. However, due to particular shifts in specific NGAs, the patterns are much less smooth.

In relation to one another, what sociopolitical norms did various societies around the Mediterranean and Asia evolve over time?

Moralizing norms, and to a lesser extent, prosociality promotion, the belief that rulers were not gods, a formal legal code, general applicability of law, constraints on the executive, and full-time bureaucrats were features shared by all NGAs except Kansai, possibly due to the NGA’s high value of null values.

Galilee, the Kachi Plain, the Konya Plain, Susiana, Upper Egypt, and Crete(to a lesser extent) performed significantly better in terms of equating rulers and commoners with elites.

The Yellow River Valley did have a marginally greater instance of impeachment than Upper Egypt.

With the exception of Kansai, every NGA has a much significant number of full-time bureaucrats than Latium.

Moralistic punishment and belief in omniscient supernatural beings were prevalent throughout the Classical world, with the exception of Cambodia and the Kachi Plain, respectively.

Crete had a common set of detailed and formalized rules, but individual welfare was emphasized more than in Cambodia. Though rulers may be replaced in certain cases, there was a significant disparity in the perceived importance of rulers and commoners. There was also a belief in an omniscient god at times.

Galilee had a highly structured legal code, which was commonly imposed and bureaucratized, but with a much stronger focus on community good and moralistic punishment. Equality was given a higher priority, and executives were subjected to further restrictions. Since belief in an omniscient supernatural being was prevalent, rulers became less likely to be called gods. And then, once in power, rulers are scarcely replaced by bureaucratic means.

Over time, the Cambodian Basin was defined by a culture that prioritized the community over the individual, as well as a formalized, broadly applied legal system that supported an advanced bureaucracy. There seems to have been a lot of inequality, although there were few restrictions on power relations.

On the Kachi Plain, society was less likely to punish you for religious rather than legal purposes, despite the intense pressure to abide by social moral norms. Though laws were a little replaceable, rulers, elites, and commoners had surprisingly equal inherent value in the eyes of society.

Social pressure to fit in was high in Kansai, which, combined with a formalized and widely enforced legal code and a powerful bureaucracy, likely meant that people generally understood what was required of them in society. Despite the fact that elites and commoners were relatively equal, the ruler was God.

The profile of the Konya Plain was similar to that of Galilee, with some rule of law, a focus on the public good, moderate equality, a restricted executive, and a belief that rulers were not divine, as well as a belief in an Omniscient supernatural being. Moralistic punishment, on the other hand, was less widespread than in Galilee.

The Middle Yellow River Valley, like Cambodia and Kansai, exemplified a combination of prosocial norms, a solid legal system, and a bureaucracy. However, rulers were not always regarded as divine and could be institutionally restrained at times. Since elites were less equated with commoners in the Yellow River Valley, social stratification may have been higher than in Kansai.

Latium exemplified the rule of law, with standardized legal codes preferred over moralistic punishment, executive restrictions, and even a strong history of impeachment. The rulers were not divine, and there was widespread belief in an all-powerful supernatural being. However, inequality was widespread, with commoners usually not perceived as important by society as elites or rulers.

Despite clear prosocial and moral standards, a structured legal code, and an army of bureaucrats, laws seem to have been less widely implemented in Susiana. Executives faced moderate constraints, and social stratification was minimal.

In Egypt, moralizing norms were of high importance, though a formalized and general legal framework existed. There was a strong bureaucracy and equality between elites and commoners was relatively high. Although rulers were at times considered divine, they could be highly institutionally constrained.

Which of these norms were most likely to be observed at the same time and in the same place?

A formal legal code, prosociality promotion, and general applicability of law were the most likely to be observed together quantitatively, with a clear possible correlation in the results. Impeachment was the least likely trait to co-occur with equating elites and commoners, all of which had a mild negative association.

Impeachment was also correlated to the involvement of full-time bureaucrats, though the connection was much weaker.

Impeachment and an omniscient supernatural being, the promotion of prosociality, and moralistic punishment were the least correlated traits, implying that the latter three did not correlate with the presence or absence of the former in any substantive way.

Axial age was a pivotal time in early human history when human beings began to reflect for the first time about individual existence, and the meaning of life and death.

Could an Axial Age Happen Again?

Some say that we may be on the verge of a new one now. There is no doubt that technology has changed the way people live their lives, connect with society, communicate, and view the world around them, both individually and collectively. The first axial age marked a discovery of transcendence. Let’s see what the future holds.

Connect with me on LinkedIn here.

--

--

Bharti Prasad

Data Science Research Intern at Indian School of Business | Kaggle Expert | IIITDM-Jabalpur’22