Key concepts
The Mortality of States Index (MOROS) provides an overview of the lifespan of different political states for the past 5,000 years. It documents commonly agreed state formation and end dates for over 440 different political states, from 3100 BCE (Egyptian Dynasties I and II) to 2021. We define a state as a set of centralized institutions that coercively extract resources from, and impose rules on, a territorially circumscribed population. Mortality in this case refers to the end or termination of that political unit: we have different datasets coded for slightly different definitions of termination, ranging from simple commonly agreed historical periodisation through to the loss of sovereignty (political independence) for 25 years. For now, to the best of our knowledge, MOROS is the largest dataset of state lifespans.
Purpose of the tool
MOROS is a dataset that can be used for finding patterns in the longevity, persistence, spread, and fragility of political states. The key inputs are the most reputable estimates of state formation and termination dates. Alongside this, MOROS is also being updated with a listing of the most compelling and cited causes of collapse/transformation for each state.
MOROS has been used as the basis for an analysis of state longevity that is set to be published in the Proceedings of the National Academy of Sciences (PNAS).
We recommend that the tool is used carefully. Users should bear in mind that many of the listed dates are uncertain and often based on provisional historical and archaeological evidence. Moreover, the current version does not separate collapses in which a significant amount of the population and capital was lost for a prolonged period of time over a substantial territory, with simple losses of sovereignty, changes in ruling elites, or transformations in government. Future versions will look to code for this by using expert elicitation. Users should be careful not to confuse the entries (states) with other political units such as societies or ‘civilizations.
History and background
This tool originated from a desire to examine whether there was a pattern in the rise and fall of societies. There was no clear database of state lifespans, and sources such as Wikipedia and Encyclopaedia Britannica are not peer-reviewed and often inconsistent. A much earlier, preliminary version was used in the article ‘Are We on the Road to Collapse’ for BBC Future in 2019. After publication, I realised that there was substantial room for improvement by drawing on more sources and using different coded versions of the data.
From 2020-2022 I worked on developing a new, updated version. This was driven by an ongoing project analysing whether states age (grow increasingly vulnerable and likely to terminate over time).
The initial data source in 2019 was three different surveys of historical empires by Rein Taagepera. These were then complemented by:
- The Seshat Database.
- The four-volume 2016 Encyclopedia of Empire.
- The Correlates of War Project.
We then added 22 additional profiles from speciality sources, primarily on Chinese dynasties and Korean kingdoms. 30 entries were excluded due to high uncertainty over dating, or uncertainty as to whether they constituted a political state. This constitutes the largest ‘maximalist’ version. In 2022-2023 we created a ‘mega-empires’ set of the 59 empires that have exceeded a territorial area of 1 million square kilometres. We also recorded the data with a more strict and consistent definition of state termination as the loss of sovereignty and political independence for more than 25 years (roughly a generation). This is the ‘minimalist’ version. For comparison, we have also have datasets of estimates from Wikipedia and Encyclopaedia Britannica.
We have several intended approaches for improving this dataset:
- Using expert elicitation to add in missing polities and verify dates.
- Using a structured review of the collapse literature to add details on the proximate cause of state termination, the primary theories of why the state ended, and coding whether it was a collapse, transformation, conquest, or dynastic shift.
Future plans
MOROS will result in the following outputs over the coming year:
- A paper titled ‘The Vulnerability of Aging States: A survival analysis across pre-modern societies’ has been accepted by PNAS and should be published in the coming months.
- A manuscript titled “The Mortality of States (MOROS) Dataset” will be submitted to Nature Data by the end of April. This will include a summary of MOROS, the methods used to create and validate the data, as well as text and xml copies of the data.
- The book “Goliath’s Curse: A Deep History and Future of Societal Collapse” (to be published by Penguin in early 2024) will draw on MOROS and the societal ageing paper.
Where to get started
The MOROS manuscript in Nature Data will provide a useful summary of MOROS, its background, how it was created and validated, how it should be used, as well the data in both text and xml formats.
The Supplementary Information in ‘The Vulnerability of Aging States: A survival analysis across pre-modern societies’ paper includes information on how to use the dataset for a survival analysis, as well as a link to the code that was used.
For further information, people should contact Luke Kemp - luka.kemp@outlook.com
CSER's 3 top tips for using MOROS
- Be aware that these are estimates of state formation and termination dates. Not all of these were ‘collapses’ or disastrous events.
- Be cognizant that MOROS includes multiple datasets (including the minimalist and maximalist versions, as well as the Mega-Empires subset) and make sure to use all of the datasets when conducting an analysis. This can help ensure that your findings are nuanced, robust, and may reveal important findings (such as different survival rates for mega-empires compared to states more broadly).
- Contribute to MOROS. MOROS is intended to be a live, updatable tool. Hence, we encourage users (especially specialists) to submit corrections and further data that may be missing. This will be verified by a member of the MOROS team and then uploaded to the main versions.