Competitive Pluralism
Competitive pluralism is the institutional feature Goldstone's Why Europe? identifies as the most distinctive structural advantage of Northwestern Europe in the early modern period. Europe was not unified. It was fragmented into dozens of competing states — kingdoms, principalities, city-states, republics, ecclesiastical territories — none of which could establish continental hegemony. This fragmentation was in many respects inefficient: the wars, the duplicated administrative structures, the incompatible legal systems. But the fragmentation produced a structural incentive for innovation that no unitary empire could match. A state that suppressed a useful innovation lost competitive advantage to neighboring states that adopted it. A state that persecuted skilled minorities lost those minorities to rivals eager to attract them. A state that imposed excessive taxation drove capital and talent across its borders. The competition was a sorting mechanism that rewarded institutional openness and punished institutional closure, not perfectly or immediately, but persistently over centuries.
The Infrastructure of Fragmentation — Contrarian ^ Opus
There is a parallel reading that begins from the material substrate required for competition itself. Competitive pluralism in early modern Europe wasn't merely a matter of political boundaries — it rested on a specific geography of rivers, mountain ranges, and coastal access that made continental unification prohibitively expensive. The Rhine, the Alps, the Pyrenees weren't just lines on maps but physical barriers that required massive energy expenditure to overcome. This natural fragmentation created the competitive dynamics Goldstone celebrates, but it was an accident of geology, not a reproducible institutional design.
The AI moment reveals the substrate dependency starkly. Digital infrastructure knows no Alps. The concentration of computational resources, training data, and technical talent in a handful of locations — Silicon Valley, Shenzhen, perhaps London — occurs because the physics of computation rewards co-location and the economics of data favor aggregation. The competitive pluralism that Europe enjoyed was premised on the high cost of conquering territory; the consolidation of AI capabilities is premised on the low cost of transmitting information. When a frontier model can be deployed globally from a single data center, when training requires datasets that span continents, when the talent pool is so specialized that all roads lead to the same few labs, the structural conditions for pluralism have evaporated. We are watching the construction of digital empires that would make the Ming dynasty envious — unified not by imperial decree but by network effects, economies of scale, and the sheer capital requirements of staying competitive. The question isn't whether we can preserve competitive pluralism through regulation, but whether competitive pluralism is even possible when the means of production are inherently centralizing.
In the AI Story
The contrast with Imperial China is the central comparison. China, unified under a single imperial administration, could suppress innovations deemed threatening to social order without facing competitive pressure from neighboring states of comparable power. When the Ming dynasty decided in the early fifteenth century to dismantle Zheng He's fleet and prohibit oceanic exploration, no competing Chinese state could exploit the withdrawal. The decision — not irrational within the logic of a unitary empire concerned with internal stability — was catastrophic for the continuation of Chinese commercial and technological leadership. It closed an entire domain of innovation without any external pressure to reverse the decision. European states, facing each other as competitors, could not make comparable withdrawals without handing advantage to rivals.
The mechanism operates in subtler ways as well. Religious dissidents persecuted in one European state could flee to another that welcomed them. This mobility constrained persecution and distributed the benefits of skilled migration. The Sephardic Jews expelled from Iberia in 1492 found refuge in Amsterdam, where their commercial networks contributed to the Dutch Golden Age. The Huguenots expelled from France in 1685 brought their textile expertise to the Netherlands, England, and Prussia, enriching their host societies at France's expense. These flows were not aberrations. They were the normal functioning of a competitive state system in which persecuting minorities meant losing them to competitors.
The same dynamic operated in intellectual and scientific domains. Galileo's conflict with the Catholic Church is remembered precisely because it was exceptional. The broader pattern was that heterodox thinkers could find institutional homes across a fragmented Europe. Descartes could move from France to the Netherlands. Spinoza could work in Amsterdam. The Royal Society of London, founded in 1660, created institutional protection for empirical inquiry that was copied by academies across Europe. No single political or religious authority could suppress scientific inquiry across the continent, which meant inquiry could accumulate across centuries despite episodes of local suppression.
Applied to the AI moment, competitive pluralism is partially present and trending toward concentration. The global technology market is competitive at the state level — the United States, China, and the European Union represent meaningfully different regulatory and commercial environments, and capital and talent flow among them. But within each of these jurisdictions, AI development is concentrating in a small number of firms with the capital, data, and computational resources required to train frontier models. The concentration of AI means is structurally concerning: if the competitive dynamics that once operated among many commercial actors collapse into competition among a handful of frontier labs, the incentive for institutional openness that competitive pluralism provides weakens. The broligarchs critique gains analytical weight.
The question for sustained AI bloom is whether the institutional openness that competitive pluralism produces can be preserved through antitrust enforcement, open-source AI development, and international regulatory arbitrage. If the AI industry consolidates into oligopoly, the structural incentive for institutional openness weakens. Firms with monopoly positions do not need to tolerate heterodox ideas or attract skilled migrants. They can extract rents from their position without facing competitive pressure. The loss of competitive pluralism at the industry level is one of the most concerning structural trends of the current moment, and it occurs silently — not through dramatic political events but through the quiet consolidation of capability in the firms that can afford frontier model development.
Origin
The concept builds on a long tradition in comparative political economy, including Eric Jones's The European Miracle (1981) and the work of Charles Tilly on state formation. Goldstone's contribution is to integrate competitive pluralism with the other institutional features of the Why Europe? framework, demonstrating that no single feature was sufficient and that the combination was historically rare.
Key Ideas
Fragmentation as advantage. Apparent inefficiency of multiple competing jurisdictions produces structural incentives for openness that unitary authorities cannot replicate.
Exit as constraint. Skilled migrants can flee persecution; capital can flow to friendlier jurisdictions; innovations can be adopted by rival states. These exit options constrain persecutors.
Contrast with imperial unity. Song China's unitary administration could suppress disruptive innovation; no comparable suppression was possible in fragmented Europe.
AI-era application. The mechanism still operates at the state level among the US, China, and EU; it is weakening at the industry level as AI development concentrates.
Debates & Critiques
Scholars have debated whether competitive pluralism was genuinely causal in European development or whether it correlates with other factors (geography, Protestant Reformation, specific resource endowments). The counterfactual cannot be tested directly, but comparative analysis suggests that unified empires — Ming China, Mughal India, Ottoman Empire, Imperial Russia — systematically exhibited the institutional closures that fragmented Europe avoided. The mechanism's operation in specific episodes (religious refugee flows, scientific institutional diffusion, commercial innovation adoption) is well documented, which provides indirect support for its causal role.
Appears in the Orange Pill Cycle
Layers of Competition — Arbitrator ^ Opus
The tension between these views dissolves when we recognize that competitive pluralism operates at different scales simultaneously, with varying intensities at each layer. At the level of sovereign states, Edo's framework remains 90% correct — the US-China-EU triad creates genuine competitive pressure that prevents any single regulatory philosophy from achieving global dominance. When the EU mandates algorithmic transparency, American firms must decide whether to comply, exit, or fragment their offerings. This is classical competitive pluralism operating exactly as Goldstone described it in early modern Europe.
At the infrastructure layer, however, the contrarian view captures 80% of the reality. The physics of computation and the economics of data do create powerful centralizing forces that didn't exist in fragmented Europe. A handful of firms control the compute clusters necessary for frontier model training; even fewer control the specialized chip fabrication that makes such training possible. But — and this is where 20% of Edo's optimism remains justified — open-source movements, academic research, and the surprising efficiency gains in smaller models create pockets of genuine competition even within this concentrated landscape.
The synthetic frame that emerges is one of nested competitions with different equilibrium points. Perhaps the right question isn't whether competitive pluralism will survive in its traditional form, but what new institutional arrangements might preserve its benefits — the incentive for openness, the constraint on persecution, the reward for innovation — within a technical landscape that tends toward concentration. The European fragmentation that Goldstone studied was partly geographical accident, yes, but it was sustained by institutional innovations like the balance of power and international law. Our task may be to develop equivalent institutions for digital civilization — antitrust adapted to network effects, international agreements on compute access, or novel forms of technological federalism that maintain competition without sacrificing the efficiencies of scale.
Further reading
- Jack A. Goldstone, Why Europe? The Rise of the West in World History 1500–1850 (McGraw-Hill, 2008).
- Eric Jones, The European Miracle: Environments, Economies and Geopolitics in the History of Europe and Asia (Cambridge University Press, 1981; revised 2003).
- Joel Mokyr, A Culture of Growth: The Origins of the Modern Economy (Princeton, 2017).
- David Cosandey, Le secret de l'Occident: vers une théorie générale du progrès scientifique (Flammarion, 1997).