Rise of nations: Why do empires expand and fall?

S. A. Vakulenko, Dmitry Lyakhov, Andreas Weber⋆, Dmitry V. Lukichev, Dominik L. MichelsChaos 2020 · DOI

TL;DR

This paper models "empires" as centralized networks with a few dominant "centers" and many "satellites," using a "divide and rule" strategy. They find that such empires are most stable with a single center and can achieve complete control over their dynamics by adjusting center-satellite interactions, but this control comes at the cost of slow development, leading to a "control trap" for large empires.

What Problem Does This Paper Address?

The paper investigates the dynamics and stability of complex systems that exhibit hierarchical structures, drawing parallels to historical empires and modern organizations. Specifically, it addresses how centralized control, implemented through a "divide and rule" strategy, affects the stability and evolutionary potential of these systems, particularly in the presence of noise and internal dynamics. The authors aim to understand the conditions under which such "empires" can maintain stability, exert control, and adapt to challenges, and why they might eventually fail or stagnate. This is relevant to understanding phenomena ranging from political and economic systems to biological networks.

Methods

The authors employ a mathematical framework based on a continuous-time Hopfield neural network model. They adapt this model to represent "empires" with distinct "center" and "satellite" nodes. The core of their approach involves:

  • Divide and Rule Model: Defining interactions where centers strongly influence satellites, but satellite-satellite interactions are weak or absent. This is mathematically represented by specific forms of interaction matrices (A, B, C, D).
  • Stochastic Stability Analysis: Investigating the probability of the system remaining in a stable state over time, considering the impact of random noise (stochastic fluctuations). This involves calculating probabilities like $P_{sur,cent}$ and $P_{sur,net}$.
  • Inertial Manifold Theory: Applying concepts from dynamical systems theory to understand how the complex dynamics of a large system can be effectively governed by a smaller set of variables (the centers), especially when centers evolve slowly.
  • Numerical Simulations: Using simulations to explore the temporal evolution of empire activity under various conditions, including different types of noise and satellite interactions.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • Single Center Stability: Empires with a single ruling center are found to be more stable than those with multiple centers.
  • Control via Interactions: The "divide and rule" framework allows for complete control over the empire's dynamics by adjusting the strength of interactions between centers and satellites.
  • The "Control Trap": While centers can exert control, this requires them to evolve slowly. For large empires, this slow evolution leads to a "control trap" where development becomes extremely slow and eventually stagnates, hindering adaptation and growth.
  • Noise and Stability: The nature of noise correlations significantly impacts network stability. Negative noise correlations favor globally connected networks, while positive correlations favor structures with bounded connectivity.
  • Bose-Einstein Condensation Analogy: The concept of Bose-Einstein condensation is used to explain how certain network structures can be maximally stable against noise, particularly when noise correlations are negative.
  • Maximal Flexibility: Divide and rule networks can be "maximally flexible," meaning they can generate a wide range of dynamical regimes, including periodic and chaotic behaviors, by tuning center-satellite interactions.

Key Results

  • Structural Stability: Empires with $n$ centers and $N > 1$ satellites can generate all structurally stable dynamics of dimension $n$, controllable by center-satellite interactions (Theorem 2).
  • Optimal Network Topology: For positively correlated noise, a single center with maximal connectivity is most stable. For negative correlations, bounded connectivity is optimal.
  • Slow Dynamics and Control: Total control is possible only when centers evolve slowly. This leads to a "control trap" where increasing size hinders evolution.
  • Numerical Evidence: Simulations show that empires are generally stable against uncorrelated noise but vulnerable to correlated noise. Satellite mutualism can enhance recovery after a crisis, while competition can diminish stability.

Limitations and Future Work

  • Idealized Model: The model is an idealization, particularly the "kleptocratic regime" where satellite viability is initially ignored. Real-world systems are more complex.
  • Noise on Centers: The analysis primarily focuses on noise affecting satellites. The more complex case of noise affecting centers is postponed for future work.
  • Strong Satellite Interactions: The paper focuses on weak satellite interactions. The behavior of empires with strong satellite interactions is not explored.
  • Future Directions: The authors suggest exploring the impact of noise on centers, investigating strong satellite interactions, and further analyzing the "control trap" phenomenon in the context of economic growth and technological progress. They also note that the concept of inertial manifolds is crucial for understanding the dynamics of large dissipative systems.

Summary updated

Abstract

We consider centralized networks composed of multiple satellites arranged around a few dominating super-egoistic centers. These so-called empires are organized using a divide and rule framework enforcing strong center-satellite interactions while keeping the pairwise interactions between the satellites sufficiently weak. We present a stochastic stability analysis, in which we consider these dynamical systems as stable if the centers have sufficient resources while the satellites have no value. Our model is based on a Hopfield type network that proved its significance in the field of artificial intelligence. Using this model, it is shown that the divide and rule framework provides important advantages: it allows for completely controlling the dynamics in a straight-forward way by adjusting center-satellite interactions. Moreover, it is shown that such empires should only have a single ruling center to provide sufficient stability. To survive, empires should have switching mechanisms implementing adequate behavior models by choosing appropriate local attractors in order to correctly respond to internal and external challenges. By an analogy with Bose-Einstein condensation, we show that if the noise correlations are negative for each pair of nodes, then the most stable structure with respect to noise is a globally connected network. For social systems, we show that controllability by their centers is only possible if the centers evolve slowly. Except for short periods when the state approaches a certain stable state, the development of such structures is very slow and negatively correlated with the size of the system's structure. Hence, increasing size eventually ends up in the "control trap."
BibTeX
@inproceedings{vakulenko2020rise,
  author = {S. A. Vakulenko and Dmitry Lyakhov and Andreas Weber⋆ and Dmitry V. Lukichev and Dominik L. Michels},
  title = {{Rise of nations: Why do empires expand and fall?}},
  booktitle = {Chaos},
  year = {2020},
  doi = {10.1063/5.0004795},
}

Loading related content...

Figure