To that conflict of four hundred years we owe the rise of civil liberty. If the Church had continued to buttress the thrones of the king whom it anointed, or if the struggle had terminated speedily in an undivided victory, all Europe would have sunk down under a Byzantine or Muscovite despotism. For the aim of both contending parties was absolute authority. But although liberty was not the end for which they strove, it was the means by which the temporal and spiritual power called the nations to their aid.

Lord Acton

1 Introduction

In this paper, I am revisiting an issue that I have dealt with in Møller and Skaaning (2014, Chapter 8). My general conclusion remains much the same, but the arguments have been specified—indeed, a number of them are completely new—and I sift a much broader body of historical evidence to probe these arguments.

In recent decades, the rule of law has been singled out as a factor conducive to increasing human welfare, spurring economic growth, and creating political stability (Belton 2005, p. 5; Carothers 2006; Haggard et al. 2008, pp. 205–234; cf. Tamanaha 2004; Møller and Skaaning 2014). However, the rule of law has also proven to be very difficult to develop—much more difficult than, say, democracy (Møller and Skaaning 2013). Hence, the proliferation of what has been termed “delegative” or “illiberal” democracies, that is, political systems in which electoral competition for power occurs in the absence of the rule of law and liberal freedom rights (O’Donnell 2007; Zakaria 1997, pp. 22–43; See Møller and Skaaning 2013). It is against this backdrop that scholars have revisited the historical origins of the rule of law (Acemoglu and Robinson 2012; Blaydes and Chaney 2013, pp. 16–34; Fukuyama 2011, 2014; Stasavage 2016).Footnote 2 This new literature in many ways marks a return to—and draws on—a large “classical” scholarship that rotated around the Western route to the modern state and modern democracy (see Møller 2017). This classical research agenda has been captured by phrases such as the “The Rise of the West” and the question “Why Europe?” (Chirot 1985, pp. 181–195; Hall 2001, pp. 488–497).

One point of consensus spans the new and the older literature: A set of norms and institutions that arose in Western and Central Europe in the Middle Ages were a harbinger of what later became the rule of law. The British economist Jones (2008[1981], p. xiv) has captured this sentiment in his claim that “Europe alone managed the politically remarkable feat of curtailing arbitrary power” (see also Stasavage 2016). Not everyone would agree with Jones’s categorical statement about European uniqueness (see, e.g., Goldstone 2000, pp. 175–194). That said, it is uncontroversial to point out that medieval Europe was saturated with the notion of law and what we would today call constitutionalism (Downing 1992; Møller and Skaaning 2014; Tamanaha 2004; Tierney 1982; Berman 1983). There is also a general consensus that, in this guise, the rule of law preceded modern state building and political and economic “modernization” by centuries (e.g., Finer 1997, p. 870; Fukuyama 2011; Sabetti 2004, pp. 70–78).

What, more specifically, are the medieval norms and institutions that provided a harbinger of the rule of law? Whereas many modern rule of law conceptions address the problem of unruly power, the relevant medieval notions concerned the arbitrariness that follows from unlimited power (Krygier 2016, pp. 209–210; Tamanaha 2004). As Berman (1983, p. 9) puts it, the gist of the idea was that “law transcends politics”. Law thereby became “disembedded”, something that stood above and constrained rulers (Berman 1983, p. 86). This is often rendered as the notion of the supremacy of law, meaning that the ruler was not a lawmaker but bound by law (Fukuyama 2011; Møller and Skaaning 2014; Tamanaha 2004). This finds expressions in medieval sayings such as “law was found not made” and that it was “a matter of knowledge rather than of will” (Tierney 1982, p. 30). These norms were embodied in institutions such as charters of liberties and corporate rights more generally. Returning to the normative level, these institutions were legitimized by theories about limited monarchy, rule based on consent, the right to resistance, and corporations (Bloch 1971[1939], pp. 451–452; Tierney 1982).

A number of scholars have made the case for this medieval legacy facilitating later world-historical developments such as democracy, the modern state, and economic affluence (Acemoglu et al. Acemoglu et al. 2001, pp. 1369–1401, 2002, pp. 1231–1294, 2008, pp. 808–842; Downing 1992; Jones 2008[1981]; Møller 2015, pp. 110–123, Møller 2017). However, both the more recent and the older literature have had a tendency to treat the precocious European development of norms and institutions pertaining to the rule of law as something exogenous. That is, sparse attempts have so far been made to analyze the causes that brought them about in the first place. Part of the problem seems to be that recent social science literature has not engaged carefully enough with the work of medieval historians.

The purpose of this article is to review both what the social science literature has to say about the medieval origins of the rule of law and to bring the work of medieval historians into this discussion. The article proceeds as follows. First, the phenomenon to be explained is identified. Second, the article discusses what so far seems to be the major “origins” hypothesis, namely, that the norms and institutions pertaining to the supremacy of law were a contingent product of the secular-religious conflicts in the High Middle Ages, reinforced by the rediscovery of Roman law and the political theory of Aristotle. This is contrasted with an alternative hypothesis, which traces these developments deep into the Early Middle Ages. In this connection, the matter of the quality of historical sources is briefly touched upon. On this basis, an attempt is made to bridge the two competing interpretations of the historical evidence.

2 The Secular-Religious Divide

The first thing to note is a widespread consensus about the “proximate” cause. A large literature agrees that the development of the norms and institutions described above was associated with—indeed, possibly sparked by—the medieval divide between religious and secular power (see particularly Berman 1983). As Tierney (1988, pp. 1–2) points out, this duality is the most exceptional feature of the medieval West and one that left its mark on both societal institutions and normative political theory. This disjunction worked to spread the notions of the supremacy of law in a number of ways.

The premise of this influence is the bare salience of the medieval Church (Southern 1970, pp. 15–16). As Finer (1997, p. 857) points out, “medieval Europe was saturated in Christian values”. It is difficult to exaggerate the extent to which the Church provided the common mental frame of reference in a society that was preciously short on genuine state power (Mann 1986, Chapter 12). The Church was the one organization that was virtually ubiquitous in European societies. Indeed, it reached into even the smallest parish. Moreover, the Church had a monopoly on education and learning, at least until the medieval universities began to break free from ecclesiastical structures (Berman 1983, pp. 161–162; Howe 2016, 204–229). The Church therefore had the normative power to influence conceptions and institutions in medieval Europe (Mann 1986, 376–379). This is where the conflict between church and state comes to the fore. For this conflict triggered a series of large-scale attempts by the Church to constrain secular rulers—reinforced by an attempt by secular rulers to fortify their own legitimacy at the expense of the Church.

First, the Church contributed to the balance of power that has historically characterized the European multistate system (see Møller 2014, pp. 660–670). Most spectacularly, it actively hindered the holy Roman Emperors from achieving a European hegemony in the eleventh, twelfth, and thirteenth centuries (Southern 1970; Tierney 1988). In a more general sense, the Church attempted to hinder any secular ruler from gaining pre-eminence over the others by supporting the maxim Rex in regno suo imperator (Ullmann 1970[1955], 454). That is, monarchs were each other’s equals, not beholden to a higher secular authority such as that of the emperor (or of a king-of-kings) (Hall 1985; Ullmann 1977, pp. 49–50). This was necessary because the medieval Church did not have the material means (most importantly, the armies) to partake in the political power struggle of Western Christendom. Its only important sanction was excommunication, and to achieve its political aims, it needed the “consent and cooperation of secular rulers” (Southern 1970, p. 20). Only a multistate system with vertical relationships between political units could ensure that such cooperation would be forthcoming on a regular basis.

Second, the Church denied secular rulers religious power and splendor, including power over the ecclesiastical infrastructure. By the late first millennium, secular rulers had achieved a “quasi-sacerdotal splendor”, and they held sway over ecclesiastical investiture (Ullmann 1970[1955], p. 232; Southern 1970, p. 34; Berman 1983, pp. 88–91; Howe 2016, pp. 232–235). Throughout Western Christendom, monarchs were seen as leaders of the Church in their territories, they convened church councils to deal with religious issues, and they appointed bishops, abbots, and occasionally even priests. In the second half of the eleventh century, the Church denied them these rights and stole their religious thunder.

Third, the Church followed up the material work of balancing powerful secular rulers with an equally important normative influence. The Church actively spread the notion that secular rulers were bound by Christian law (Finer 1997, pp. 863–864; Hintze 1975[1931], p. 318). This notion was integrated directly into Gratian’s collection of Canon law (Berman 1983, p. 145). This presented the background for theories of limited monarchy and the mixed constitution (Tierney 1982). Moreover, the Church spread the notion of corporations in general and corporate representation in particular, thereby legitimizing rule by consent (Berman 1983, pp. 146–149). These notions represented a way of securing the immunity of the clergy from secular power and of giving the clergy a voice in the secular body politics. But they also derived from theological and political debates about the organization of the Church itself (Tierney 1982, 1988). These theories were later transplanted into secular political theory where they presented the justification for the notion of political representation and constraints on rulers (Tierney 1982, pp. 11–25). An important consequence of the standoff between lay rulers and the pope was thus “the growth of doctrines justifying resistance to unjust rulers”, first within the Church and subsequently within secular polities (Tierney 1988, p. 86; Berman 1983).

Fourth, the conflict over investiture gave an important impetus to law and litigation. The secular rulers and the Church fought each other over investiture and jurisdiction (Tierney 1982; Berman 1983). The result was that law came to regulate conflict and that law was constantly refined. The eleventh century rediscovery and subsequent application of Roman law was to a large extent an attempt by, first, the Church and, later, secular rulers to find judicial arguments to buttress their positions over investiture and jurisdiction (Ullmann 1970[1955], pp. 367–368). As Max Weber observed, it was the trial of strength between the Catholic Church and the monarchs that paved the way for an independent corporate stratum of lawyers and for secular law (Collins 1986, p. 113). The profession of jurists emerged “in response to the need to reconcile the conflicts that raged within the church, between the church and the secular authorities, and among and within the various secular polities” (Berman 1983, p. 160). More generally, the Church attempted to make the notion of law ubiquitous in Western Christendom because law presented a weapon that it was more skilled at wielding than the monarchs were (Berman 1983, p. 95) (Table 1).Footnote 3

Table 1 Consequences of secular-religious divide relevant for the rule of law
Full size table

3 The Gregorian Revolution

So, how can we explain the medieval disjunction between church and state? The most common answer in the literature is to point to the so-called Gregorian Revolution or Gregorian Reforms in the second half of the eleventh century. The magisterial exposition of this thesis has been made by Berman (1983) in Law and Revolution. Berman points out that a tremendous transformation took place in legal systems between 1000–1050 and 1150–1200, what he more specifically refers to as the “Papal Revolution of 1075–1122” (15–19). Berman clearly sees this as a contingent breakthrough, orchestrated by reform popes who broke with everything that went before. Until then, the clergy had supported the imperial concept (67). This was mirrored in a form of ceasaropapism (often referred to as Rex-Sacerdos) that according to Berman was broadly similar to that found in the Byzantine Empire (88–91). The result of the papal revolution was that “Law became disembedded” (86). More particularly, Berman’s (1983, p. 50) principal thesis is that the primary impulse behind Western legal systems—or the Western rule of law—owes to the eleventh century “assertion of papal supremacy”.

More recently, this thesis has been endorsed by Francis Fukuyama (2011) in his work on the origins of political order. The starting point for Fukuyama’s analysis is similar to the starting point of this article, namely, that medieval Europe was idiosyncratic by having developed factors pertaining to the rule of law in the absence of state capacity. Fukuyama seeks the origins of this development in the Gregorian Revolution. Like Berman (1983), he construes this as a contingent event, spearheaded by a series of prominent ecclesiastical actors, including Pope Gregory himself.

The story about the Gregorian reformers begins in 1049 when Leo IX was elected pope.Footnote 4 Paradoxically, his election was forced through by German Emperor Henry III with the explicit intention to reform the corrupt Roman Church (Ullmann 1970[1955], pp. 251–252).Footnote 5 Leo died in 1054 but was succeeded by a string of other reform popes. In 1059, a papal decree established that the cardinals were henceforth to elect the pope, an attempt to secure the independence of the Church from both German emperors and the Roman nobility (Ullmann 1970[1955], p. 298; Tierney 1988, p. 36). The conflict escalated in the 1070s, after Gregory VII threw down the gauntlet to German Emperor Henry IV, the son and successor of Henry III. The result was the so-called Investiture Conflict over who had the right to invest especially bishops with their offices. In line with the other reform popes, Gregory took a stern line on clerical celibacy and simony (clerical corruption). However, Gregory went further than his predecessors in challenging secular influence over the Church. Most famous is his so-called Dictatus Papae, 27 statements of powers arrogated to the pope, which was published in 1075. The aptly named papal bull Libertas ecclesiae of 1079 further solidified Gregory’s position. These doctrines spelled out the notion of the pope’s right to depose monarchs (“That it may be permitted to him to depose emperors”) and the ecclesiastical right of investiture (“That he alone can depose or reinstate bishops”). The Investiture Conflict followed as Henry IV refused to abide by these doctrines.

Southern (1970, p. 103) holds that the Dictatus Papae had much greater political effects than the Magna Carta; Berman (1983, p. 96) points out that the document itself was “revolutionary”. The conflict over investiture was so heated because monarchs could not well yield to the papal demands. The appointment of bishops was simply too important a means of political and economic patronage, and the bishops were a crucial part of medieval government (Southern 1970; Tierney 1988, p. 86; Ullmann 1970[1955], p. 296). Indeed, according to Ullmann (1977, p. 33), no king in eleventh century Western Europe could govern his realm without the support of the higher clergy in that area. Hence, what followed was a centuries long, acrimonious standoff, in which both papacy and monarchs resorted to judicial arguments, based on Canon law, the rediscovered Roman law, and ultimately, the political thoughts of Aristotle. In this way, the Investiture Conflict was a forceful impetus behind the increasing importance of law and litigation and the notion that law bound rulers (Southern 1970; Tierney 1982, 1988; Berman 1983, p. 111).

Berman’s (1983) and Fukuyama’s (2011) analyses of the Gregorian Revolution resemble the notion of a “critical juncture” that triggers a path-dependent sequence (Mahoney 2000, pp. 507–548; Capoccia and Kelemen 2007, pp. 341–369). Both clearly construe the events of the late eleventh century as ones that were not bound to take place, indeed, would not be expected to take place based on deeper structural variables. Rather, they see them as a contingent product of actors’ choices, most notoriously those of the reform popes in general and Gregory VII in particular. Next, they emphasize that these actors’ choices were to have fundamental consequences for the subsequent European development as they set the Western part of the continent onto a completely new path. This understanding of the Gregorian reforms is illustrated in Fig. 1.

Fig. 1
figure 1

The Gregorian reforms as critical juncture

Full size image

It is possible to back this point of view with some rather simple empirical observations. Here is how Southern (1970, p. 96), in his classical work on the medieval Church, summarizes the relationship between religious and secular power in the period 700–1050.

The affairs of the church received little direction from Rome. Monasteries and bishoprics were founded, and bishops and abbots were appointed by lay rulers without hindrance or objection; councils were summoned by kings; kings and bishops legislated for their local churches about tithes, ordeals, Sunday observance, penance; saints were raised to the alters – all without reference to Rome.

Already, the Merovingian kings had legitimized themselves through religious power (Fouracre 2005, p. 381; Ullmann 1970[1955], p. 150). However, as Southern (1970) points out, by the tenth century kings had gained a genuine theocratic role, including the right to invest bishops with offices (Tierney 1988, p. 25). This development had taken place in a context in which the papacy was extremely weak. Even Ullmann (1970[1955]), ever keen to stress the patristic roots of the Gregorian reforms (more on this below), observes that in the late ninth and early tenth centuries, the German Emperor had made the church an intrinsic part of his imperial government, and that both emperor Otto III and emperor Henry II were at one and the same time emperor and ‘pope’ (Rex-Sacerdos) (244). Indeed, Henry III directly assumed power to appoint popes (259–262), a making and unmaking of popes that was broadly similar to the Byzantine emperors’ appointment of patriarchs (322). This relationship was then, Southern (1970, p. 34) observes, turned up-side down in the span of a single lifetime.

The social and religious order which has just been sketched showed little sign of breaking up in the year 1050…And yet within the next sixty or seventy years the outlook had changed in almost every respect. The secular ruler had been demoted from his position of quasi-sacerdotal splendour, the pope had assumed a new power of intervention and direction in both spiritual and secular affairs…

The notion that the eleventh century disjunction was not bound to take place can be further buttressed through a comparison with church-state relations in Eastern or Orthodox Christianity. Here, we find similar religious doctrines and traditions inherited from early Christianity. We also find an important common legal and political inheritance from Rome. Just like medieval Western Christendom, the Byzantine Empire was law-bound. Indeed, by the eleventh century, it was law-bound in a much more sophisticated way than the polities in Western Europe, based on the Roman law that was only about to be rediscovered to the West (Finer 1997, p. 629, pp. 649–655). It is all the more striking that we find no disjunction between religious and secular powers in Eastern Christendom. On the contrary, the Byzantine—as later the Russian—relationship was one in which there was no crisp distinction between powers as church and state was supposed to be in symphony, or at least in harmony (Ostrowski 2006, pp. 224–225). Throughout Byzantine history, the emperor was in charge of a theocracy: he had an intrinsic religious legitimacy and presided over church affairs, meaning that the Church did not have the independent position we find in the West (Runciman 1977, pp. 162–164; Kazhdan and Constable 1982, pp. 94–95). This position was later inherited by the Russian tsars. Hall (1985, pp. 1119–1120) notes that in the Orthodox Church, the very same Christian doctrines that were invoked by the reform popes were used to legitimize the notion that religious and secular power should be collapsed into one autocratic office, what above was termed Rex-Sacerdos or ceasaropapism. Others have added that in the Byzantine Empire, religion was much more conservative and intolerant than we see it in Western Christendom (Finer 1997, p. 661).

4 An Earlier Disjunction?

Much thus seems to support the notion that the Gregorian Revolution is the key to the medieval development of the rule of law. However, it is not cut in stone that this revolution was indeed so revolutionary. The notion of a critical juncture is technically that actors make choices between alternatives that cannot be predicted based on prior events and social, economic or political structures (Mahoney 2000, pp. 507–548). The potential objection is that the events of the eleventh century were endogenous to deeper structural factors (see Møller 2012, pp. 693–715). To understand whether this was the case, one needs to further probe relevant features of early medieval Europe.

Here, we encounter a rather arduous obstacle, namely, that the centuries before 1000 AD are to a large extent shrouded in darkness with respect to the historical evidence. In terms of narrative sources, what we have are copies of copies, with the later copies often containing later additions, meaning that the original text is difficult to discern (Halsall 2005, p. 61). Moreover, the narrative evidence that exists often concerns monarchs, nobles, or the higher clergy, meaning that we know very little about, for instance, institutions and activities on the local level in this period (Reynolds 1997[1984], p. 88). Of course, we have archaeological findings, but these are difficult to interpret for those wishing to understand legal or political norms and the workings of legal and political institutions.

Hence, any discussion about early medieval society and political institutions is apt to be somewhat vague and tentative. That said, we can make a number of important observations, the first being that the Gregorian reformers emphatically claimed to base their doctrines on Christianity in general and earlier papal doctrines in particular. In fact, the eleventh century reform popes saw their project as an endeavor to rescue the old and true ways of a Church that had gone astray (Ullmann 1970[1955]; Tierney 1988, p. 47).

Next, it is obvious that long before Gregory VII and Henry IV, there was at least a latent competition between religious and lay rulers in Western Christendom. Indeed, “[t]he possibility of a continuing tension between church and state was inherent in the very beginnings of Christian religion” (Tierney 1988, p. 7). The conventional view is that the cleavage between secular and religious power was first activated by Pope Gelasius (pp. 492–496), who in the fifth century formulated the doctrine of papal supremacy over secular rulers (Cantor 1993 [1963], pp. 86–87, p. 177; Ullmann 1970[1955], pp. 20–21). The doctrine of the two swords—the secular potestas and the religious auctoritas Footnote 6—is normally attributed to Gelasius. This doctrine was emphatically a response against the Byzantine emperors’ attempt to rule the Church, and it implied that there was no lay jurisdiction over clerics (Ullmann 1970[1955], pp. 20–27).

However, Ullmann (1960, p. 25) has forcefully argued that the “fixation of papal primacy” took place a generation earlier, under Leo I (pp. 440–461). Furthermore, Ullmann argues that this—as opposed to the later Gregorian reforms—was a true critical juncture as there was little in prior papal literature “to facilitate his [Leo’s] argumentation”. Leo’s rupture consisted in arguing that the pope possessed Petrine jurisdictional powers, based on the so-called Petrine commission (28). It was not that Leo invented this idea—which is based on Matthew 16:18—but it was he who constructed the doctrine of the pope as the legal “heir to St. Peter” (33). This doctrine was so powerful because, according to the Roman law of the day, there was no difference between the heir and the deceased, meaning that the pope received Peter’s “plenitude of power” (34–36). Ullmann (1960, p. 46) points out that the “medieval papacy was built on the juristic foundations laid by Leo” (46). Indeed, Ullmann (1960, p. 49) specifically argues that it is an

incontrovertible fact that the fully fledged papal-hierocratic theory in the Middle Ages had firm roots deep down in the patristic age: in ideological substance there was no difference, the only difference was that later theory—so largely Leonine theory—was applied in practice.

Ullmann (1970[1955]) adduces a number of examples to the fact that what he terms the “papal-hierocratic” idea was thriving long before the Gregorian reforms. For instance, in the ninth century Nicholas I and Adrian II further propounded Gelasius’s notion that the pope was superior to the emperor (197–204). See from this vantage point, what the reform popes achieved in the eleventh century was “the implementation of the hierocratic tenets, that is, the translation of abstract principles into governmental actions” (262). The hierocratic doctrine, going all the way back to Leo I and Gelasius I in the fifth century, now became “the governmental basis of the papacy” (271). Indeed, Gregory’s excommunication of Henry IV was, Ullmann (1970[1955], p. 283) observes “based upon Gelasian authority”. Even the model for cardinal elections of popes had a historical foundation in the form of the “old Roman decree of 769 passed in the synod under Stephen III” (322), and ultimately in the canonical elections of bishops in the early Church (Monahan 1987, pp. 43–47).

More generally, the entire idea of an independent Church found a basis in the writings of the Western Church fathers of Late Antiquity. Augustine’s City of God was an attempt to construe the Church as something living outside of the secular realm, meaning that its destiny was not bound up on that of the Western empire, which was failing.Footnote 7 As such, there was an important early difference in the doctrines expounded in the West by e.g. Augustin and those expounded by churchmen in the East (see also Chadwick 1967; Runciman 1977).

It can therefore be argued that we can identify an early disjunction between church-state relations in Western and Eastern Christendom (see also Tierney 1988, pp. 9–10). Most importantly, there was a conspicuous difference between the religiously backed title as emperor in the West and in the East. In the East, the emperor was elected by epiphany, in the West he was invested with his office by the pope (Finer 1997, p. 630; Tierney 1988, pp. 16–17). Indeed, according to Southern (1970, p. 60), the ninth century popes made the Carolingians their political clients by crowning them as emperors, thereby reversing the “Eastern” hierarchy between emperor and church (see also Ullmann 1970[1955], p. 67; Tierney 1988). Furthermore, we can note that several ninth and tenth century popes—most prominently aforementioned Nicholas I (pp. 858–867) and his successor Adrian II (pp. 867–872)—protested against the monarchs’ right to investiture, albeit they were too weak to change things (Tierney 1988, p. 25; Berman 1983, p. 93). Likewise, clerical celibacy and canonical election “belonged to the early tradition of the Western Church” (Tierney 1988, p. 47) even if these practices had largely lapsed by the tenth century. Especially canonical election was important because it meant that the notion of political consent could be backed by references to practices in the early church (Monahan 1987, pp. 43–47). Finally, the forgery of the Donation of Constantine—which popes used as a lever to claim power over Western monarchs after the Gregorian Revolution—also dates back to the centuries before AD 1000, probably to the eighth century. As Ullmann (1970[1955], p. 81) points out, the Donation made the pope a Papstkaiser and “from here to Gregory VII is only a small step”. Furthermore, the subsequent great forgeries of the ninth century—e.g. the so-called Pseudo-Isidore—did not as such create new ideas but rather attempted to give already accepted ideas a historical foundation (Ullmann 1970[1955], p. 188). The notion of historical reproduction is illustrated in Fig. 2.

Fig. 2
figure 2

The Gregorian reforms as a product of deeper historical factors

Full size image

Of course, we can also find examples of conflict between churchmen and lay rulers in Eastern Christendom. For instance, Byzantium was regularly riddled with conflicts between the patriarch and the emperor (see, e.g., Runciman 1977). Similar examples can be adduced from Orthodox Russia. The most famous here is probably that of Patriarch Nikon in Moscow, who held the office from 1652. Nikon pronounced that the Church was superior to the secular rulers, and even attempted to act on the basis of such preeminence (Riasanovsky 1969, p. 200). However, Nikon’s fate is telling for church-state relations in Eastern Christendom. In 1666–1667, he was deposed by a church council. More generally, in the East, the ability of the Church to restrain rulers depended upon personalities, in particular, the personality of the head of the Church, exactly because the doctrine of symphony gave no institutional basis for ecclesiastical opposition to lay rulers and, hence, to political constraints.

This is where we find a large difference between Western and Eastern Christendom. The Gregorian reformers had more of a tradition to invoke and were much better placed institutionally to oppose secular rulers. Finally, the very fact that Western Christendom was characterized by a nascent multistate system made it possible for popes to play a divide-and-conquer game that was not available to Eastern churchmen. This was the way the popes tried to weaken the great Hohenstaufen emperors in the twelfth and thirteenth centuries (Munz 1969). Hall (1985) provides another telling example. In 1312, Holy Roman Emperor Henry VII requested that the pope order Robert of Naples to go to the imperial court at Pisa to face charges from Henry. Pope Clement V refused and instead published the bull Pastoralis Curia, which stated that a king is sovereign in his own realm, not under the authority of the emperor. Thus, the Church publically recognized the earlier mentioned doctrine of Rex in regno suo imperator (see also Ullmann 1970[1955], p. 454).

We can also make a somewhat different observation, which underscores that the medieval institutions and norms pertaining to the rule of law did not appear out of the blue in the eleventh century. As Reynolds (1982, pp. 15–16) points out, even in the Early Middle Ages, lay political ideas were based on the supremacy of law, and custom and towns—to the extent that they existed—were self-governing collectives. To quote her on the situation in England:

All sorts of groups were allowed to act in a way that later centuries would associate with corporate status. Every community had its customs, which seem to have involved some hazy right of legislation – the making of by-laws in terms of the later English law of corporations (Reynolds 1997[1984], 35).

Reynolds (1997[1984], p. 5) goes as far as to argue that the rediscovery of Roman law and of Aristotle and the codification of canon law did not make much of a difference with respect to political ideas about and practices of collective groups. Rather, these were rooted in an earlier tradition that goes back to the period before the Gregorian Revolution. To put it differently, these new judicial conceptions and political theories simply clarified what already existed on the ground (Reynolds 1997[1984], pp. 64–68).

5 An Integrated Explanatory Framework

Though the reform popes, including Gregory, did not have to invent their position de novo, some of their reforms had no actual precedent.Footnote 8 For instance, this was the case for the right of the pope to depose monarchs (Tierney 1988, p. 56). There is little doubt that the reform popes in general and Gregory in particular here went further than earlier papal doctrine and church tradition. It is also pertinent to recall that in the Middle Ages, authority was stronger the older it was. This goes a long way toward explaining why the reform popes did everything to show the “ancientness of the ideas put forward and applied” (Ullmann 1970[1955], p. 360; see also Berman 1983). In the centuries to follow, this led to a theory of papal supremacy that came very close to justifying a new theocratic role, no longer for the secular rulers but for the popes (Southern 1970; Tierney 1988).

However, vis-à-vis the situation in Eastern Christendom, it is clear that the Western environment was more propitious for a rivalry between religious and secular power. The case can therefore be made for a synthesis that stresses that a set of deeper structural factors paved the way for what was admittedly revolutionary action by the eleventh century reform popes. This conclusion can be situated in-between two of the most important works reviewed: While Berman (1983) goes too far in arguing for the existence of a ceasaropapism similar to that of the Byzantine Empire prior to the “papal revolution”, Ullmann (1970[1955]) goes too far in stressing the historical continuity of the “papal-hierocratic” doctrines from patristic time to the High Middle Ages.

The analytical purchase of such an integrated explanatory framework can be illustrated by considering the political and legal inventiveness of the High Middle Ages. It was within the church that the notion of the corporation was transferred to cover a larger community, in the first place that of the Church. The notion of the corporation (universitas) was derived from Roman law but it was radically altered in the High Middle Ages. In classical Roman law it had been a concept that was purely relevant in private law. The point had been that a body of persons could organize themselves as a judicial personality. What the twelfth and thirteenth century theologians in the Catholic Church did was to argue that the Christian community, and hence the Church as an organization, was a corporation. Furthermore, they invested the notion of corporation with two aspects that it did not originally have. First, the notion of representation where the point was that the corporate group could invest an agent with plena potestas (full power), meaning that the group that appointed the agent was bound by the deals the agent struck. Second, the principle of quod omnes tangit ab omnibus approbetur (“that which affects all people must be approved by all people”), which was relevant only for a remote corner of private law in classical Roman law but which now became linked with corporate representation (Tierney 1982, pp. 23–25; Monahan 1987, pp. 99–100).

These reinterpretations of Roman law can be seen as a way for the twelfth century Church of finding ways to deal with the administrative expansion and centralization that had begun with the Gregorian reforms (Monahan 1987, pp. 112–113, p. 116). This had a series of fundamental consequences. First, it meant that the clergy, ultimately even including the pope, was bound by the laws of the corporation. Second, it meant that a majority within the Church could—via Church councils which their representative attended—decide matters that then bound everyone (Monahan 1987, p. 109, p. 256). As Berman (1983, p. 208) points out, the general councils, called by twelfth century popes, were in this respect the forerunners of the secular parliaments. However, these conciliar views never won out completely within the Church where they competed with the view that the Pope was infallible and could create doctrine (a point that was also buttressed by tenets from Roman Law, e.g. “What has pleased the Prince has the force of law’) (Black 1998). But the new political doctrines were transferred to secular political theory where representative institutions became based on the notion of the corporation. The townsmen that started to be called to royal assemblies after 1200 were thus proctors—representatives—of their town councils, which in turn were bound by the decisions made at the assemblies. Moreover, the idea that a political community could be represented as a body and that the principle of quod omnes tangit could apply to it became the very foundations of the body politics where an estate pole confronted a monarchical pole and where rule was based on consent (Poggi 1978).

Now, what is unclear is the extent to which these developments owe independently to twelfth, thirteenth and fourteenth century ideational factors—the intellectual developments within the Church—or whether they should rather be seen as normative justifications for things already practiced on the ground. As noted above, Reynolds (1997[1984]) has vigorously defended the argument that groups had long behaved in ways that were later described with the notion of corporate status. Likewise, Tierney (1982, p. 13) notes that “the Decretists lived in a twelfth-century society still soaked in the preconceptions of customary law, a society that was inclined to see law, not simply as the command of a ruler, but as an outgrowth of the whole life of a people”.

We are here touching upon an important characteristic of medieval Latin Christendom, namely, the ability to receive inputs but then reinterpret them in ways that legitimized e.g. corporate self-government (Berman 1983, p. 149). The medieval rediscovery of Roman law ultimately facilitated the creation of “sovereign, centralized states independent of papal control” (Tierney 1988, p. 98) and the rediscovery of Aristotle the “development of a philosophical theory of the state that required no appeal to theological premises” (Tierney 1988, p. 159). Roman law was pre-Christian and the lay monarchs could use it to escape the ecclesiastical control of the Church (Ullmann 1977, pp. 37–47, 1970[1955], p. 367). Indeed, as Ullmann (1970[1955], p. 367) points out, the “ideas manifested in Justinian’s codification were diametrically opposed to hierocratic ideas. For his work was the classic legal expression of the imperial-laical standpoint”. But at the same time, the Church also attempted to arrogate both Roman law and the writings of Aristotle to serve its interests. This is a neat illustration of the legal and political inventiveness of the High Middle Ages.

The purchase of this combined explanatory model can be illustrated by returning to a particular point made above: that it was not so much the rediscovery of Roman law as its medieval reinterpretation that facilitated, for instance, the notion of political representation (see also Stasavage 2016). On the basis of the integrated model, a case can be made for this owing to the interpretation of Roman law, after its rediscovery, being colored or polluted by Christianity in a way that is very different from what we find in Byzantium, where Roman law had been there first and where it never came to underpin constitutional ideas and practices.

This is to be understood quite literally. Jerome’s Vulgate Bible, dating from a period when Roman law had still been in existence, explained much in terms borrowed from this corpus. This both made it easier for medieval scholars to understand Roman law and also independently affected how it was interpreted in Western Europe (Ullmann 1977, p. 41). A body of law that construed the prince as the sole legislator thereby came to legitimize popular sovereignty through notions such as quod omnes tangit. This probably had much to do with the normative ideas about limited monarchy, the mixed constitution and representation that the Church was actively spreading.

It is similarly telling that Aristotle’s pre-Christian writings were interpreted in a way that made it compatible with Christianity. This occurred principally in the work—the so-called Summa—of the greatest medieval theologian, Thomas Aquinas. This work had a huge influence on medieval political theory. The rediscovery of the Politics was important not only because defenders of constitutionalism drew on some of Aristotle’s specific political ideas but also because it showed them that political theory could be an autonomous science (Tierney 1982, p. 29)—and that the state could be something separate from the Church (Ullmann 1970[1955], pp. 455–457). As with Roman law, this body of writings was known in the Byzantine Empire, but only in Western Christendom did it become a means for defending constitutionalism. To quote Berman (1983, p. 149):

Here again, while the Roman law of Justinian provided the basic terminology and the Greek dialectics of Plato and Aristotle provided the basic method, the combination of the two – in a whole different social context – produced something quite new.

So, the situation seems to be that the new ideas emanating from Roman law and Aristotle were refashioned to legitimize a social order that was to a large extent already in place. This is a good indication that the creative social order of the High Middle Ages owed much to what had gone before it (see Reynolds 1997[1984]).

6 Conclusions

In recent decades, the rule of law has been highlighted as the solution to numerous problems, including problems of economic underdevelopment, political misrule, and poor state capacity. This emphasis on the rule of law has made scholars probe its historical roots and, in this connection, reengage with an older literature on the modernization of the West. In the words of Alexis de Tocqueville, scholars have attempted to identify the point de départ of the European development. Surveying this literature—and perusing what medieval historians have written on the subject—two points of consensus can be identified. First, the roots of the rule of law that we know today is to be found in medieval Western Christendom. Second, the disjunction between religious and secular power is consistently singled out as a core development that sparked what I have termed the notion of the supremacy of law. More particularly, the controversies between lay rulers and popes following the eleventh century Gregorian Revolution “greatly encouraged the growth of constitutional ideas” (Tierney 1988, p. 197).

The secular-religious conflict therefore opened up “quasi-autonomous social domains” that, according to a large literature, ultimately paved the way for the modernization of the West (Holmes 1979, p. 125). In this article, I have discussed whether this disjunction was a contingent event following from the actors’ choices of eleventh century reform popes or whether it reflects a deeper historical tension between church and state in Western Christendom. The review showed that arguments can be made in favor of both of these interpretations. The Gregorian reformers did turn church-state relations upside down in the eleventh century, and some of their crucial policies had little precedence in earlier papal doctrines and church practices. However, we find a latent tension long before the eleventh century, and the reform popes saw themselves as rescuing the good old ways of the Church. On this basis, much speaks in favor of an integrated explanatory model that points out that the early medieval church-state relations in Western Christendom laid the basis for the events of the eleventh century, while still recognizing the importance of these events.

The best illustration of the consequences of this combination of deeper legacies and the events in the High Middle Ages is to be found in the peculiar flexible and dynamic interpretations of Roman law and Aristotle in the West, noted above. Rather than seeing the rediscovery of these corpuses as independent causes of the rule of law, we should see their innovative and flexible interpretation as symptoms of a social order in which factors pertaining to the rule of law could flourish. In a nutshell, both the Church and the secular rulers attempted to interpret Roman law and Aristotle to buttress their own position.

Finally, we need to answer one last question. Why did popes not reach out for the scepter of the emperors? Why did the popes not assume the theocratic position that secular rulers had had before the Gregorian Revolution? By the mid-thirteenth century, we find “a fully developed theory of papal world-government”, and several popes came very close to committing themselves to this (Tierney 1988, pp. 152–153). The simple answer is that the popes lacked the material resources to do so. First and foremost, they had little in the way of arms, instead relying on their ability to muster secular forces. Throughout the Middle Ages, the struggle between church and state therefore endured—sometimes with the pope on top, sometimes with the secular monarchs predominant. This was important because a standoff was created in which both popes and monarchs attempted to use law and litigation to have their ways.