The Feudal Origins of the Western Legal Tradition
Journal Article2

The Feudal Origins of the Western Legal Tradition

with Hilton Root

Jahrbuch für die Ordnung von Wirtschaft und Gesellschaft (Ordo), 70(1): 3-20.

Abstract. This paper draws a distinction between ‘communitarian’ and ‘rationalist’ legal orders on the basis of the implied political strategy. We argue that the West’s solution to the paradox of governance – that a government strong enough to protect rights cannot itself be restrained from violating those rights – originates in certain aspects of the feudal contract, a confluence of aspects of communitarian Germanic law, which enshrined a contractual notion of political authority, and rationalistic Roman law, which supported large-scale political organization. We trace the tradition of strong but limited government to the conflict between factions with an interest in these legal traditions – nobles and the crown, respectively – and draw limited conclusions for legal development in non-Western contexts.

The Paradox of Governance – how a government strong enough to protect its citizens against each other can be restrained from itself infringing on its citizens’ rights – is one of the most persistent problems of political philosophy. Something like a solution to the paradox, a strong-but-limited state, has been managed only in Western Europe and its descendant states. This raises several questions: first, how did Western Europe manage the feat, and second, can the solution be replicated elsewhere?

We propose to focus specifically on legal and administrative systems as the result of norms governing the relationship between a government and its citizens. By classifying such systems into the broad categories of rationalist and communitarian law, we aim to shed light on the unique trajectory of the Western legal tradition as an unlikely synthesis of elements of both, and to highlight the role of the legal system in shaping political institutions. We then show both the desirability and the difficulty of such a synthesis, and show how the development of the Western legal tradition out of Roman and Germanic law, through feudal institutions, paved the way for feasible strong-but-limited states.

Communitarian and Rationalist Law: An Overview

Analytical Approach

Scholarly frameworks for interpreting European statebuilding (e.g. Weber 1968; Ertman 1997; North et al. 2009) have typically divided the population into elites and commoners, focused on the struggles and strategies within the former group. Commoners, almost by definition, have a much broader and less coherent set of interests than political elites, and are therefore limited by higher coordination difficulties. This difficulty, the logic goes, creates space for political action by better-coordinated political elites, and justifies a focus on them (Tullock 1971). Institutional change is thus considered to be driven by the struggle among elite factions for the reins of governance.

In contrast to these approaches, we draw the dividing line differently: between a governing class and a governed population, placing the bulk of “elites” on the side of the governed.1 Contrary to a view of institutional change as a struggle among coequal elite factions, there was remarkably little turnover in the identities of the governing and governed classes over the millennium following the collapse of the Roman empire. The familial networks of Europe’s governing royalty remained mostly separate from the networks of its governed nobility (Root 2017), with the king as the nominal head, even as the balance of power between them ebbed and flowed.

Drawing the line this way allows us to focus on the norm that determines the response of the population to efforts at political domination as the basis for our analysis. This response has frequently been modeled as a collective action problem. Nevertheless, though coordination upon any particular strategy may constitute a collective action problem, coordination in general does not. Indeed, the existence of stable polities ipso facto indicates some form of coordination. This is the problem of legitimacy: political domination depends, to some extent, on the active or passive cooperation of the governed, and on their common knowledge as to which forms of domination will elicit cooperation (cf. Johnson & Koyama 2019).

The two simplest strategies will be something like “resist domination” and “submit to domination”. Both of these strategies are natural Schelling points in a political game, and both are Nash equilibria.2 If you cannot count on your fellows to resist domination, there is no incentive for you to go alone. Similarly, if your fellows are agreed to resist, there will be no rational incentive to support a would-be dominating party.3 More complex “conditional submission” strategies will be more difficult to coordinate upon, for reasons discussed below (see section 2.4). We therefore take these two simple strategies as the basis for our taxonomy.

This norm, then – whether resistance, submission, or something in between – determines the character of the legal system by shaping the problems it encounters in exercising political authority. Returning to our two baseline cases, a resisting population reliably settles into a legal structure with a particular set of characteristics that we will call communitarian (see the following subsection). On the other hand, a submissive population has the potential to scale up to the point that the exercise of political authority reliably takes on a different set of characteristics, which we will call rationalist, in order to deal effectively with the problems arising from scale (section 2.3).4 In this context, the emergence of the Western legal tradition is characterized by constant renegotiation between the two. Confronted with the exigencies of governing a population employing one strategy or the other, the governing class has less freedom to structure its administration than might otherwise be supposed. We therefore turn to a taxonomy of the strategies of the population that shape the administrative options of political authorities.

Communitarian Law

Communitarian law, the legal order resulting from a strategy of resistance, is by far the older of the two. From the very beginning of human social organization, humans have been characterized by resistance to domination and relatively egalitarian social structures (Boehm 1993). We will limit ourselves, however, to more advanced societies characterized by impersonal (but not necessarily monetary or commercial) exchange and heterogeneity in social status, and therefore some necessity for regularized law (Harwick 2018).

The primary effect of a resistance strategy is to limit the potential for power differentials to arise endogenously. This in turn limits the potential for political predation, but also prevents the formation of strong centralized governing bodies. The scale of political organization under communitarian law can sustain clan and tribal-based affiliations, but is inherently limited (Dunbar & Sosis 2018), and is likely to fracture beyond a certain size (Bandy 2004). Confederations among various populations may form – Rome fell at the hands of one such confederation – but states and empires will not. There will, for this reason, be little provision of large-scale public goods such as infrastructure or commercial law.

From the perspective of the modern developed world, communitarian law is typically referred to as “custom”. It is often unwritten and implicit in a community’s practice, and generated in response to particular internal conflicts in order to prevent them from tearing the community apart. On the one hand, this limits any dissonance between law and practice. Communitarian law is legitimated automatically, so to speak, and enforced through coordinated social pressure. It tends to be personalistic, dealing with people on the basis of their status and identity, and to resolve conflicts through consensus. Lacking the institutional capacity to raise taxes effectively, conflicts will generally be mediated informally by respected community members rather than professional judges. On the other hand, there will be no necessary congruence between the laws of separate communities, which may engender conflict when they come into contact.

In developed commercial economies, though the broader polity is governed by a rationalist framework, particular communities can nevertheless be governed by communitarian law. Examples include religious communes (e.g. Oved 1988), kibbutzim (Ben-Rafael 1997), resource commons (Ostrom 1990), and even firms (Miller 1992, ch. 10).

2.3. Rationalist law

Rationalist law, on the other hand, arises from the need to regularize and legitimize the administration of the domination that becomes possible when the governed class employ a strategy of submission, especially as the scale of political organization exceeds the feasible limits of communitarian law. There are several legal innovations that facilitate the task of large-scale administration:

  1. Recording the law in writing allows standardization of enforcement across communities, and some economies of scale in the production of trained professional judges. Leaders in a communitarian legal order may be illiterate; administrators of a rationalist order may not.
  2. Rationalist legal systems can use more abstract categories in the application of the law. Personalistic status will not necessarily be discarded as a legal concept, but it will of necessity be less fine-grained than in a communitarian legal order (cf. Johnson & Koyama 2013). This allows a wider scope for private action, simply because a bureaucracy cannot involve itself in its citizens’ lives to the same extent that the members of a single community can exercise oversight over each other.
  3. Compared with communitarian law, compliance relies more on direct compulsion administered by professionals, and less on social pressure and peer monitoring. Law is not “automatically” legitimated as in communitarian law, which allows for greater scope on the part of the sovereign for making legal changes, but also on the part of the population for lawbreaking if the law does not comport with internalized norms. If law and norms diverge sufficiently, destabilizing social movements can even be possible.

The rationalization of law is close to the notion of state capacity, which as Johnson and Koyama (2017) point out, is not, by itself, an unambiguously positive development.5 Compared to communitarian law, the primary drawback to rationalist law is the difficulty in preventing a governing body strong enough to hire professional judges and standardize the law from exploiting that power for its own benefit. The expanded sphere of private action made possible with rationalist law is a matter of administrative convenience; it is not a meaningful constraint on the governing class’ discretion to appropriate rents and suppress competing sources of power. The sovereign in a rationalist legal order, therefore, will find it difficult to make credible commitments, a fact which can impede economic development (Acemoglu 2003).

Rationalization, political centralization, and increasing social scale are driven by external (Turchin 2008) or internal (Zhao 2015) threats to the integrity of a community. Those states that were able to organize themselves in a more rationalist way were able to marshal more resources and prevail against less rationalistically ordered polities. Rome and China, for example, in the face of both pressures, both developed a highly rationalistic legal administration early in their history, systems which endure in some form or another to the present day.

2.4. Intermediate Strategies

The respective advantages and drawbacks of rationalist and communitarian law suggest the desirability of a synthesis: large-scale political organization, with safeguards against opportunistic behavior by political authorities.6 And in principle, an intermediate strategy of conditional submission would make such a legal order feasible. Under such a strategy, the population would presumptively submit to political authority, but revert to resistance in the event that leaders abused their authority in agreed upon ways. Because a rationalist legal order depends on domination not being too costly, such a reversion, if sufficiently well-coordinated, would constitute a credible threat against potential abuses of political power.

Conditional submission strategies, however, are more difficult to coordinate upon, for several related reasons.

  1. There are many separate categories of domination – economic, ideological, sexual, to name a few broad categories – not all of which will coincide. Reducing domination along one dimension, given intermediate levels of domination in others, may even entail increasing domination in others; for example, the use of political power to keep down economic inequality.
  2. In practice, therefore, the conditions that make up a coherent conditional strategy are likely to be quite complex in their details,7 and it will be difficult to create common knowledge in a large population as to which forms of domination are legitimate and which are not. A strategy that conditions its response on more information will be transmitted with less fidelity to peers and progeny, and less credible as a threat in light of imperfect information.
  3. In addition to informational difficulties, coordination upon complex strategies also faces strategic difficulties. Complex strategies are contestable along more margins, opening the door to opportunism and rent-seeking at the strategy-deciding stage. Even abstracting from (1) and (2), a simple and obviously suboptimal strategy may still be able to gather more assent than any particular complex strategy, even if any complex strategy would be a Pareto-improvement over the simple strategy.

A constitution can be thought of as a technology for coordinating upon complex strategies in order to restrain discretionary overreach by political officials. However, the track record of constitutional reform worldwide is mixed at best, and most outside a Western context have failed to meaningfully affect strategies, coalitions, or the incidence of rent-seeking (de Vanssay & Spindler 1994). Because legal norms govern the bargaining strategies among a network of concrete interest groups – in our case, between the governing and the governed – it will be difficult for a constitution, or any top-down directed change, to reconfigure an existing network of interest groups and therefore to effect meaningful changes in political strategies. In other words, because political strategies are embodied in political practice, they are likely to be “sticky” and resistant to change except over long periods of time. Where constitutions have been effective, they have been consonant with the structure of existing interest groups and built upon norms already governing the interactions among them. But constitutions themselves are neither necessary nor sufficient for this to occur.

 It remains, then, to explain the success of the Western legal tradition as a strategy of conditional submission resulting in a strong-but-limited state. Our argument is that this strategy arose during the Feudal era in Europe as interest groups favoring the rationalist Roman law and those favoring the communitarian Germanic law vied for influence and power, and eventually settled into a stable institutional compromise. We document this process in the following section, and conclude with a warning against overgeneralizing the results.

The Evolution of an Intermediate Strategy in Western Europe

The Bond of Fealty

Prior to the fall of Rome, the various Germanic tribes – the Saxons, Franks, Vandals, and Goths – were all characterized by advanced communitarian legal systems. Law was customary and unwritten, rights to land were held by families rather than individuals, and loyalty was owed to particular leaders rather than to more abstract notions of law or state.

In the aftermath of their victory over Rome, these tribes found themselves tasked with the governance of a people accustomed to a much more rationalistic legal system than their own. Desiring to shore up their own legitimacy and power, Germanic kings across Europe took on the trappings of Roman leadership, for example in issuing coins bearing their own likenesses (Duby 1974; Ertman 1997: 41). Most importantly, they issued written law codes which – while Germanic in substance – drew on categories from Roman law in exposition and organization.

One major substantive change, however, did accompany the codification of the law. Tasked with administering large numbers of foreigners, it became necessary to explicitly define the obligations between lord and vassal. Over time, this relationship came to be understood as a legal bond rather than a bond of kinship; one with definite mutual obligations, as well as formal recourse for violations. By the eleventh century it was commonly understood that in the event of a breach of fealty, the vassal could legitimately leave his lord and seek a new one (Berman 1983: 305; Bloch 1961: 451, 464).8 By the same token, if the vassal did not provide military services when called upon, or some equivalent value in money that commuted that obligation, he could forfeit his land and privileges.

The debt of Western political philosophy to this development is obvious. Berman (1983: 309) points to the bond of fealty as the source of the conception of political authority as a contract between ruler and ruled, a notion central to modern conceptions of democracy and one unique to the European legal tradition. It was also distinct from both the Roman and the Germanic sources upon which it built. Roman law was equivocal on the question of the obligations of and constraints upon rulers – a question which would occupy later jurists of public law for nearly a millennium9 – but on the whole it had a distinctly absolutist character. Traditional Germanic law, on the other hand, had ill-defined notions of contract and formal obligation, making it difficult to ensure the predictability necessary for larger scale public or private organization. Feudal law, in contrast to both, emphasized individuals as autonomous bearers of rights and obligations. And although the particular rights and obligations differed according to social and economic status, they attached to individuals in their capacity as individuals, regardless of rank.

3.2. The Codex Justinianus and the Re-Rationalization of European Law

Elements of Roman law had survived the fall of Rome in Canon law and, to a lesser extent, in the codices of the various Germanic kings. But these were compilations based on particular needs, and did not represent Roman law in its full development, which had been lost until the discovery of a full copy of the Codex Justinianus around 1050, a complete and self-contained summary of the Roman legal code commissioned by the Byzantine emperor Justinian in the 6th century.

At this time, property arrangements in feudal Europe were still recognizably Germanic. Land holding was particularly inflexible. There was no distinction between ownership and possession, rights were held to various specific uses of the land rather than to the land itself, and because these rights were held by families rather than individuals, they could only be alienated with the consent of heirs and other family members with an interest in the land. This inflexibility inhibited the movement of land into the hands of those who could use it most productively. In addition, the written law was predominantly concerned with lord-vassal relations. There was little provision either for commoners, who had recourse to the courts of their local lord, or for merchants, who had begun to establish their own independent courts.

With advances in social organization and the revival of commerce beginning to outpace the still-mostly-Germanic legal system, the advantages of an internally consistent and commercially oriented legal code over the traditional system were immediately obvious. Legal scholars across the continent were impressed by the the highly refined distinctions of the Codex – for example, between possession and ownership, or the obligations resulting from contracts versus torts – and they began in short order to establish law schools to teach the Codex (Southern 1953).

As these universities were mainly established in urban centers, which were also centers of commerce, legal learning and commercial development began to flourish in tandem. And because such cities were largely self-governing, many adopted the Codex tout court, putting them at variance with rural areas still largely governed by Germanic custom.

It was this heterogeneity across regions, in addition to more straightforward self-interest, that led the kings of Europe to push a second wave of legal rationalization. From the early 1200s, kings began to employ jurists trained in Roman law to augment their own legitimacy, and to codify and standardize custom across their domains.

In a bid to weaken feudal lords, kings at this time also began to appropriate seigneurial privileges and transform them into public functions. Most importantly, they established royal courts for serfs and commoners, and staffed them with professional judges. Manorial courts were often ineffectual, opaque, and slow, and were not provided for at all by feudal law, so royal courts extended legal protection to many who had de facto been largely without recourse to that point. As early as the eleventh century, the Church was arguing on the basis of Canon law that the relationship between a king and his subjects was governed by the duty of fealty no less than the relationship between a lord and his vassals. In this way, the emergence of an autonomous legal profession as a source of legitimacy diffused Roman law both through feudal structures and into the wider population, and provided the scaffolding for the first steps toward equality before the law in a way that would not have been possible had court systems remained the prerogative of feudal elites.10

At the same time, Roman law was hardly an unalloyed instrument of emancipation, as can be seen most clearly in France, where legal rationalization proceeded furthest.11 One aspect of the enhanced organizational ability provided for by Roman law was the organization of interest groups into corporate bodies, such as guilds and nobility, whose rights and privileges were protected by law (Root 1994: 165-68; North, et al. 2009: 69-71). The capitalization and sale of such privileges became an important source of royal revenue in France and Spain (Ertman 1997).12 The crown, for its own part, found Roman law useful in advancing absolutist legal theories and freeing itself from the constraints of law and custom. A unified state was the most natural setting for the extension of Roman law (Root 1994), and royal jurists employed its concepts to argue that the king was legibus solutus – “above the law” – and absolute in his rights as is a dominus over his estate.

This pan-European push for Romanization and rationalization continued through the seventeenth and eighteenth centuries, and – along with the diffusion of a common language of legal scholarship – resulted in what could legitimately be called a general European legal tradition.13 Over this period of time, the legal profession developed into a distinctive hub among the wider network of elites, and as an epistemic community became a crucial independent source of legitimacy in political struggles, despite the fact that local or national law in whatever form it was actually practiced and enforced was not taught in any European law school until the eighteenth century. Partly because of this persistent divergence between the theory and the practice of law, feudal privileges continued to inform the political strategies of the nobility, who increasingly found themselves at odds with kings over the latter’s use of Roman law to accumulate privileges formerly resting with the nobility.

3.3. Parliaments and the Resurgence of Communitarianism

The push for Romanization in practice was an attempt to ground royal legitimacy in the support of the people rather than the support of the nobles. The king was regarded by the nobles as a primus inter pares, but by the people as vested with supernatural authority. In addition, collective action barriers meant that any effort of the people to punish the king for malfeasance would be prohibitively difficult compared to the less numerous and better organized nobles. It is obvious why kings would prefer to establish their power on the basis of a relationship with the people, as well as why De Toqueville (1955) would argue that the nobility were the source of constraints on the sovereign.

Beginning in the 12th century in Southern Europe and moving northward in the 13th and 14th centuries (Van Zanden, et al. 2012), and proceeding in tandem with the royal push toward rationalization, nobles across the continent were able to take advantage of organizational advances in Roman law and constitute themselves into relatively permanent bodies with control over the raising of revenues for the crown. In the late 13th century, these parliamentary bodies were extended to include representatives of the cities, in recognition of their increasing wealth and influence following the post-Codex flourishing of commerce. By using their ability to levy taxes as a bargaining chip, parliaments were able – with varying success – to push back against Romanization and to safeguard their own privileges. Where the crown was able to secure an independent source of revenue, as in Spain and France, rationalization and Romanization could proceed relatively unchecked. But where fiscal necessity outstripped the king’s means, as in England and the Netherlands, the nobles were able to effectively preserve the political strategies that had developed during the feudal era and that regarded political authority as essentially contractual.

With the main power of these nascent parliaments being over the raising of revenue, many found it necessary to ally with commercial interests. Mercantile participation in these parliaments accustomed them to employ the conditional submission strategy developed by the nobility in the feudal era. In submission, the towns agreed to levy substantial taxes for the crown. But they were also able to force concessions important to the advancement of commerce, such as forbearing to debase the coinage, and more widespread protection of property rights. As Marongiu & Woolf (1968: 233) note, “in the age of absolutism, the assemblies found the strength to survive through the deep-rooted belief in the contractual nature of agreements between sovereigns and their peoples or estates.”

Especially in Northern Europe, where fiscal pressures and a lack of alternative revenue sources forced kings to call parliaments regularly, the effective restraint on royal discretion provided by parliament allowed kings – ironically – to marshal far greater fiscal resources than was possible under more successfully absolutist kings such as in France (cf. North & Weingast 1989). Towns and nobles, having received assurances that their rights were protected, were able to collect far more tax revenue than the crown’s own administration would have been capable of (Meyerson 2008). And moneylenders, assured that parliament would require the king to repay his debts, lent willingly and without the compulsion that was regularly employed by French kings (Root 1989; Veitch 1986).

Concurrent with the royal push for Romanization and the noble reaction, many jurists were also able to make cogent arguments for communitarian interests from Roman principles: not only for the rights and privileges of the nobles, but also for a sovereign constrained by law, or the holding of political authority in trust from the people. Thus armed, noble and commercial interests could oppose the crown not only on the basis of interest and power, but also on principle. By the 17th century, the interplay between parliaments and kings in Northern Europe, and especially in England, had evolved into an effectively constrained government, kept in check by (1) internal divisions within the governing class, (2) a widespread conception of political authority as contractual, which gave force to the punishment mechanisms embodied in parliament, and (3) a jurisprudence of limited government arising out of juristic debate on public law. This arrangement was further codified and rationalized in the United States, which had no nobles, but was able to emulate the internal divisions of the governing class by explicitly structuring the government into legislative, executive, and judicial branches.

Outside of Northern Europe and its successor states, much of the rest of Europe lagged behind these developments both economically and institutionally. An effectively organized resistance to full Romanization was far from inevitable: in many places, especially in Southern Europe, the sovereign was able to successfully suppress parliaments, leading to a “little divergence” in both economic growth and institutional quality (de Pleijt & van Zanden 2016). Nevertheless, the varying strength and coherence of this reaction – though leading to significant variety in European legal and political institutions – is a relatively superficial difference compared to the common legal tradition and (therefore) the conditional political strategy employed across the European continent, diffused more through the spread of an autonomous legal profession than successful resistance. The relative success of strong-but-limited governments in South and Central Europe following the two World Wars, and in Northeastern Europe following the collapse of communism, attest to the naturalness of contractual political strategies in environments with feudal heritage and a common legal language with the rest of the continent, even if those strategies had not become fully effectuated endogenously.

Lessons and Contrasts

Western Europe’s constraints on sovereign discretion derive from a legal tradition with roots in the communitarian values of the Germanic tribes. This enabled a much broader diffusion of responsibility for lawmaking and enforcement, and the conditional submission strategies that support them, than in other long-lived rationalist regimes. The obvious question then is, how can successful conditional strategies be cultivated outside of Europe, in areas without a feudal heritage?

We have sketched the path that the confluence of a rationalist and a communitarian legal tradition took toward a legal system that credibly restrains a strong central state. But there are many possible paths such a confluence can take, in principle. It will therefore be necessary to identify which aspects of the rationalization of Europe’s communitarian tradition have led to agreeable results. Given what we have argued is a relatively unified European legal tradition in spite of varying progress in its development across the continent, a sample size of one limits the definiteness with which we can answer this question. Nevertheless, the persistent failure of the constitutional liberal project to take root elsewhere in the world allows us to narrow the space of possibilities and to eliminate some obvious misinterpretations.

Consider several possible outcomes for the attempt to rationalize a communitarian legal order.

  1. The formal law imposed alongside an existing communitarian order may be simply ignored as “dead letter” – as a Chinese proverb says, “the mountains are high and the emperor is far away”. Locals may pay lip service to the formal law, but for all intents and purposes their lives remain governed by the existing communitarian law. There is little disruption, but also little possibility of exploiting the scale advantages of rationalist law. This is the case in much of Africa and Central Asia, where a state exits for the purposes of international relations, but de facto political power is mainly exercised on a tribal level.
  2. Alternatively, the rationalizing entity may impose the law at some cost, a cost which rises with the variance between formal law and existing practice. To the extent that existing practice has been shaped by the necessity to reduce recurring conflicts, imposed rationalization is likely to resurface old conflicts (Leeson & Coyne 2012), which the rationalist law may or may not have sufficient provision for, even if people eventually acclimate to it. The privatization of land holdings by colonial governments in certain African tribes, for example, cleared away many of the communitarian defenses against predation and empowered the headman to exercise a great deal of arbitrary authority that had not been previously possible (Van de Walle 2001).
  3. Finally, communitarian law may be rationalized endogenously, either de novo or from the attraction of an external example. Such was the process in post-Roman Europe, where Germanic kings slowly adopted pieces of Roman law – first only the legal categories to aid in the articulation of the existing law, but over time more and more substantiative pieces driven by the needs of the contemporary society.

In addition to the problem of constructing a stable conditional political strategy, these possibilities point to the existence of a parallel problem: the legitimacy of any change in political strategy. Any equilibrium strategy will require, first, the common knowledge that others are deploying it,14 and second, the incentive compatibility of deploying it given that knowledge.

Any sharp break in political strategy – as for example in (2) – will, by necessity, involve a break in common knowledge, and for that reason a collapse of the legitimacy of any strategy (cf. Pejovich 1999: 171). The process of converging back upon an equilibrium strategy may be fitful, violent, and given local conditions, may or may not result in one that is compatible with the imposed legal order. On the other hand, slow change – though more amenable to the maintenance of legitimacy – can prolong contestability, and result in worse equilibrium strategies than even a break in legitimacy, as ‘gradualists’ have learned in the wake of the collapse of Communism. The maintenance of a trajectory of change over longer periods of time may itself not be time-consistent (Roháč 2013).

Third, in the era of nation-states, nearly the entire world is governed, at least in principle, by rationalist legal systems. The example of Western Europe and its successor states stands as a strong rationalizing attractor for nearly every culture in the world. Unfortunately, its example shines no light on the reverse problem of “communitarianizing” a rationalist legal order, except that perhaps collapse is necessary – and even so, there must be a viable communitarianism waiting to pick up the slack. China, for example, extirpated its communitarian traditions very early in its history in response to the threat from the Mongol steppes. Despite numerous collapses over the course of its several-millennia history, however, there has never arisen a legal force akin to the feudal tradition to insulate its citizens from an overbearing central state.15

In a sense, the lessons of the evolution of Europe’s legal order are not especially actionable, and complicate any efforts to replicate its success elsewhere. Even so, some broad principles are generally applicable.

First, legitimacy matters. Political strategies are embedded in practice (Harwick 2019), and generally depend for their stability on a particular persistent pattern of bargaining among interest groups. Convergence is a hard-won process. It follows that such norms will be resistant to change imposed from the outside, for better or for worse. Detailed and highly individualized efforts will be necessary to minimize normative disruption during the process of legal change, and a workable plan in one part of the world is not likely to be adaptable elsewhere.

One important implication is that it is important to avoid a legal “vacuum” during which short time-horizon rent-seeking becomes rational and competing legal and political strategies rush in. In France, for example, seigneurial courts continued to hear cases until well into the era of monarchical absolutism. By contrast, in many newly independent countries, especially former colonies, new legal regimes were manufactured and transported from the colonist to replace the older customs that governed traditional communal relations. Communal custom was eliminated before a viable alternative had come into play, leaving many aspects of community activity without legal safeguard. This vacuum enabled the local “big man” to expropriate many of the rights of the population and use connections with the center to cement those usurpations. The imposition of national law enabled those with access to it to act without constraints and with limited responsiveness to local populations, rendering the new institutions of the state into tools of oppression.

Second, legal changes are more likely to be legitimated when they address existing problems experienced by the people subject to them. Medieval kings got purchase on the establishment of Roman law courts because of the unsatisfactory provision of justice by manorial courts. Cities adopted the Codex Justinianus in whole or in part because of the difficulty of governing commercial relations under Germanic law. Germanic kings adopted Roman elements in response to the difficulty of administering vastly expanded territories. Even the initial rationalization of Rome’s legal system was initiated by the demand of the plebeian class for written rules in response to a sense of unfair and self-serving rulings, and developed over the course of the next millennium through jurists’ discussion of contemporary legal problems (Stein 1999, ch. 1). Any project of directed legal change, therefore, will have a greater chance of success the more it can point to specific dissatisfactions with the existing legal system and offer a credibly superior alternative.

Finally, especially in more advanced stages of political organization, legal strategies affect political outcomes primarily via their effect on the organized coalitions who employ them. In Europe, the confluence of Roman and Germanic law resulted in a stable balance of power between the crown and the nobles that was eventually formalized into recognizably liberal governance structures. In the U.S., the same political strategy made artifactual political coalitions – namely, a bicameral legislature and an executive – viable and stable. Such coalitions will not be stable (at least in the intended form) without a congruent political strategy, as the adoption of US-like constitutions across the world with little success shows.


The legal tradition of Western Europe and its successor states stands alone in the world as a system that enables a great deal of political and organizational complexity, and at the same time credibly restrains the exercise of political power enough to support a market economy with significant concentrations of private wealth. This system has been the backbone of the Western world’s “great divergence” over the past several centuries, one which has supported the production of historically unprecedented wealth. That a rationalist theory of rights could be established from out of feudal loyalties gives the Western legal tradition a unique character not found in other current or past regimes.

The connection of that system to economic growth has been well-studied, from the very beginning of the discipline of political economy to the more recent explosion of interest in the economics of institutions. The origin of particular features of that system has been comparatively neglected, especially given the problem of striking the right balance between narrowly documenting the historical development and broadly theorizing about the entire space of possibilities in light of the single extant example. Approaching the development of the Western legal tradition using the ideal types of ‘rationalist’ and ‘communitarian’ legal systems brings attention to an important but often overlooked element of the European legal tradition, and also has broader applicability outside Europe. We have made some tentative forays into drawing broader lessons; it remains for future work to apply the distinction elsewhere.


  1. Though it does provide additional insights, our framework does not necessarily rule out a flat-elite view of European statebuilding if the governed elites act as normative focal points among the broader governed population.
  2. More generally, these strategies are also evolutionarily stable, meaning their stability and viability do not depend on the rationally optimizing behavior of the agents who employ it.
  3. A strategy of resistance does not have to imply realized costly behavior on the part of each member of the resisting public; only that each member signals a willingness at some margin to bear the cost of resistance. The more such people as a proportion of a population, the more credible the threat against would-be dominators, and the fewer resistors will in fact have to bear any costs. On the credibility of such signals in light of the incentive for preference falsification and the potential instability of the resulting equilibria, see Kuran (1995).
  4. Similar dichotomies have been offered by numerous authors. Greif (1994)’s “collectivism” and “individualism” are perhaps the most similar, though they govern strategies in a mercantile game characterized by principal-agent problems where politics are very much in the background. Other more explicitly political distinctions employed in a European context, such as Weber’s (1968) “patrimonialism” and “bureaucracy” or Ertman’s (1997) “authoritarianism” and “constitutionalism”, refer specifically to variation among post-Feudal European states. All of these latter distinctions, therefore, will be more or less orthogonal to the rationalist-communitarian distinction.
  5. “State capacity”, however, is defined in terms of outcomes (“the ability of a state to collect taxes, enforce law and order, and provide public goods”) rather than as an independent feature. For this reason, depending on the particular measure used, empirical work using it as an independent variable is therefore subject to the Glaeser et al. (2004) critique. Legal rationalization, as an independent feature, avoids this critique.
  6. Djankov et al’s (2003) Institutional Possibilities Frontier construction depicts this relationship with a convex curve representing a tradeoff between private predation and public predation. Contra Djankov, however, the argument in this section is that there is no automatic mechanism to move a society to a cost-minimum intermediate position, and in fact the poles (corresponding to communitarian and rationalist law, respectively) will be attractors on the curve.
  7. The debate between the Federalists and the Anti-federalists during the ratification of the U.S. constitution is an instructive illustration of this point.
  8. This was a key feature distinguishing European feudalism from, for example, Japanese feudalism, where “the vassal’s submission was much more unilateral” and “the divine power of the Emperor remained outside the structure of vassal engagements” (Bloch 1961: 452).
  9. At various points the great Roman jurist Ulpian had declared both that “the sovereign is not bound by the laws” (D.1.3.1) and that “it befits the majesty of the ruler to profess that he as emperor is bound by the laws” (C.1.14.4) (Stein 1999: 59). While it is likely that Ulpian had intended the latter statement more as a practical expedient than a principled limit on imperial authority, medieval jurists with communitarian intuitions used it to give force to the feudal notion of mutual political obligation.
  10. North et al. (2009) argue that equality before the law originates as a way of structuring and pacifying inter-elite conflicts, and is later generalized to the masses. By contrast, this account draws attention to the fact that certain rationalistic aspects of equality before the law were first extended to the masses by the expansion of royal power, with the despotic excesses of legal rationalization held in check by the reaction of feudal elites. See the following subsection.
  11. Rationalization in this sense (i.e. the regularization of obligations under the law, where France excelled) is distinct from bureaucratization (i.e. the professionalization of administration, where France notably lagged England and Germany).
  12. This practice was a major complaint of the revolutionaries in the French Revolution, and in fact did not cease until the Code Napoléon supplanted France’s mix of Roman and feudal law.
  13. This is not to diminish the variety in post-feudal European political organization. Much attention has been focused on constitutional England versus absolutist France, for example. Even so, France and England have much more in common than, say, England and Egypt, and it will be these commonalities upon which the paper focuses.
  14. In more advanced stages of political and social organization, it will be organized coalitions rather than individuals employing the relevant political strategy. In such societies, the organized coalition abstracts from heterogeneity in the level of political motivation among its members and the various separate coordinated responses made necessary by an attempt at domination.
  15. Technical feasibility, however, limited the influence that the central state could influence for a long time. In addition, the Confucian legal tradition consisted in strong norms of self-regulation and good governance (Zhao 2015). Even so, there existed no independent power base to challenge the emperor’s – or the party’s – power should Confucian norms fail to hold sway. This is especially true now that technological advances have removed many of the practical barriers that had formerly protected some individual autonomy (Devereaux & Peng 2018).


Game theoryHistoryInstitutionsLawPolitical EconomyRule of LawDoug NorthGordon TullockMark KoyamaMax WeberNoel JohnsonThomas ErtmanTimur Kuran


Facebook Twitter Reddit Threads


  • 1


    Apr 01, 2020 at 11:35 | Reply

    There was an anthropologist named Marvin Harris back in the day who discussed feudalism. His assessment was that having so much access to farmland and rain made it possible for small tribes/communities to be more true to themselves, rather than getting absorbed into empires (at least for long).

    • 1.1

      Cameron Harwick

      Apr 01, 2020 at 11:53

      Makes sense – I guess the argument would be, submitting to political authority sucks, and you’ll only be willing to do it voluntarily if there wouldn’t be enough resources otherwise? Kinda the flipside of the Hydraulic Empire argument.

Leave a Reply

More Content