Multilevel Selection II
MLS in Cultural Macroevolution
In the previous post on this topic, The Controversy over Multilevel Selection, I discussed multilevel selection (MLS) in general, as well as its application to solving the puzzle of Major Evolutionary Transitions. But for my research program the most interesting application of MLS is in Cultural Macroevolution.
Cultural macroevolution (CME) is a subfield of cultural evolution that studies large-scale changes in the cultural traits of whole groups. A central question in CME is how and why characteristics of polities (such as chiefdoms, states, and empires) evolve over time. Over the past couple decades I found that the application of MLS to cultural evolution provides a very useful theoretical framework for understanding the evolution of states and empires over the past 10,000 years.
We need a multi-level approach because many polity-level characteristics evolve under selection pressures that act in opposite directions at different levels of hierarchical organization. For example, a provincial ruler and the elite network that provides the basis for his power, may decide that it is in their interest to secede from the overall kingdom. They may dislike taking orders from the king and to be subordinated to the elites in the capital. Even more importantly, they would prefer to keep the taxes, collected in their province, for their own uses rather than share them with the center. When all, or most, dukes decide to secede and form their own independent duchies, the king will not be able to coerce all of them. Such territorial disintegration happened a myriad of times in history.
So, why do we have large states and empires? In fact, the overall trend during the past 10,000 years has been with larger polities replacing small ones. The answer is that large polities, everything else being equal, are stronger than small ones, and they tend to win in the inter-state competition.
The overall direction of evolution, thus, will depend on the balance of centrifugal and centripetal forces. As this balance shifts, in either direction, we see either an integrative trend to larger states, or a disintegrative trend to smaller ones. This is precisely the focus of MLS, and dynamical models based on this theory.
Let’s dig a little deeper into this question. Large empires did not evolve simply as a result of a village growing in population and territory to a huge size. Instead, they are an evolutionary result of repeated combinations integrating smaller-level units (interspersed with periodic disintegrations). Thus, simple chiefdoms united several villages, complex chiefdoms united several simple chiefdoms, and early states united several complex chiefdoms. Territorial organization of large states and empires—hierarchical nesting of territorial units such as provinces, districts, and localities—often reflects this evolutionary history, just as biological organisms retain many traces of their evolutionary history.
A great example of such iterative evolution of social scale is the history of the region that is today France. In the Iron Age, on the eve of the Roman conquest, Gaul was inhabited by a great number of tribes (chiefdoms), most of them belonging to one or other tribal confederation (complex chiefdoms). We know a lot about this political landscape from Julius Caesar, who conquered all of them.
After the conquest, the Roman organization of Gaul simply reflected this tribal organization. Romans divided Gaul into c. 300 pagi (singular pagus, which became the French pays), corresponding to the territories inhabited by individual tribes. The Roman pagus became a county during the Carolingian period and a bailiwick in medieval and early modern France. A Carolingian count (comes) supervised the viscounts (viscomes), just as the later bailiff (bailli) had prévôts as his subordinates.
Under the Romans, the pagi were grouped into civitates, corresponding to the tribal confederations. There were about 60 of them, greatly varying in size. However, they were dominated by 10 larger confederations (the Arverni, the Aedui, the Santones, etc.), which together made up half of Gaul. The Roman civitates roughly correspond to the later provinces/gouvernements. The number of these administrative units varied over time, but on the eve of the revolution, there were 40, including 7 very small ones. This is not a bad match to the 60 Roman civitates since eighteenth-century France occupied roughly two thirds of the Roman Gaul territory. Several modern French provinces still retain the names of the pre-Roman tribal confederations (e.g., the Arverni—Auvergne, the Santones—Saintonge, the Bituriges—Berri). Others (e.g., Brittany, Normandy, and Burgundy) were named after later invaders.
Territorial integration of France, which happened repeatedly between the Iron Age and the early modern age, always occurred in steps, in which smaller-scale units were aggregated into large-scale units, which, in turn, were aggregated into even larger units. In other words, integration proceeded in a hierarchical manner.
Disintegration, similarly, was a multistep process. Thus, when the Carolingian empire collapsed in the ninth century, it was first divided into larger-scale units—France, Germany, and Lotharingia. Next, France disintegrated into duchy-sized units (e.g., Burgundy, Aquitaine, and Provence), which, in turn, fragmented into pagi/counties. In some regions, the process of disintegration went even further, and counties fissioned into castellanies.
The reintegration of France under the Capetian kings was similarly a lengthy and laborious process because the Capetians had to start at the lowest level from their base in Île-de-France, reducing castles one by one and installing in them castellans loyal to the dynasty. Similar processes were occurring elsewhere in France at that time.
After bringing counties under their control, the counts attempted to expand their power over a duchy. In some cases, they adopted ducal titles (as in the Duchies of Aquitaine and Burgundy), but in other cases they continued to call themselves counts, even though they ruled over duchy-sized territories (as in the counties of Flanders and Champagne). Later on, most of these duchy-sized territories became provinces of the French kingdom.
The French example is a particularly nice illustration of how the process of integration (and dissolution) of large territorial states. But all complex large-scale societies evolved by this general mechanism, and that’s why they are organized in many hierarchical levels. And that’s why MLS is such a productive theoretical framework to study such processes.
If you are interested to further pursue these ideas, take a look at my recent writings:
Your framework on inter-polity competition and free-riding is compelling, but I think the causal chain sometimes gets compressed too much: war and frontier pressure explain selection (who survives and expands) quite well, but they explain less well why some systems become self-complexifying administrative orders while others do not.
Free-riding exists everywhere, but its prevalence and destructive force depend on the incentive regime. As long as expansion (a functioning frontier) yields reliable returns—booty, status, new positions in the hierarchy—it is usually more profitable for elites and the coercive apparatus to stay “in the line” than to tear the system apart from within. In that phase, free-riding exists, but it remains bounded: it is easier and safer to take a share of expansion than to privatise internal flows.
The switch happens when the frontier, for various reasons, stops paying for itself—when it yields less than it costs in risk and expenditure. Then the military machine changes its source of nourishment: external growth gives way to consuming the internal base. Precisely at that point, free-riding stops being a marginal deviation and becomes mass behaviour, because internal predation becomes more rewarding than external service.
From there, the outcome depends less on war per se and more on whether the centre can reconfigure internal resource management and elite organisation. That requires: (1) building regular internal revenues and circulation instead of living off windfalls; (2) maintaining a controlled form of elite competition—strong enough to produce managerial improvement, but not so open that the upper stratum is flooded by claimants to rent; and (3) making peaceful life sufficiently rewarding (together with security) that a critical mass of the population does not treat collapse as an acceptable outcome. If this reconfiguration fails, centrifugal forces begin to exceed the integrative force of the centre.
So frontier pressure and war matter as an external filter, but “growth of complexity” emerges only where military success can be converted into internal protocols: regular revenue extraction, reproducible administration, disciplined elite selection, mechanisms that block mass access to rent, and a broadly beneficial order for the majority. Without that conversion, very large empires are possible that never become self-developing systems of governance.
And the hardest question is this: what happens when everything has already been taken, the frontier disappears, and the system is left alone with itself?
At that point, the external outlet for ambition closes. If a significant share of competitive energy was previously spent outward (war, new lands, booty, new statuses), it now inevitably turns inward. From there, only two stable trajectories are available.
Either the empire manages to make internal life the new “frontier”—creating internal sources of growth and status that are more rewarding than destruction: regular fiscal capacity instead of plunder, expansion of internal markets and exchange, controlled elite competition, advancement channels based on competence and service rather than sheer seizure, and a sufficiently rewarding peace (plus security) such that the majority experiences order as a gain.
Or it fails—and then the disappearance of the frontier becomes a generator of centrifugal dynamics: ambition feeds on internal resources, the elite stratum swells and privatises flows, the apparatus turns obligations into private “feeding rights,” and the population stops seeing a meaningful difference between “order” and extraction. In that case, collapse does not require an external enemy: it is produced by the incentive structure itself, because the system did not find a replacement for the frontier as a mechanism for distributing ambition and rent.
Societies have broad inequalities in many things besides the things we normally associate with the word inequality. Body mass, political polarization, facial features, cognitive skills, access to status mobility, but the range of these inequalities have only increased with the industrial revolution, almost as if there is some energy-mediated conservation of volatility that extends outside the domain of material wealth to encompass all other social domains.
One analogy that comes to mind are the increasing amplitudes of nodes and antinodes in a harmonic wave on a drum as the energy increases. Has this been encountered elsewhere by you, perhaps under different terms?